Re: [sqlite] Suggests for improving the SQLite website
John Stanton wrote: How about having adding a social networking capability so that non-technical people will have a reason to use the website. You cannot expect to attract them with a frugal and highly functional embedded database library. LOL! - Richard - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
Great idea! Why don't we give them little printable chits for free chips and beer as well?! Just the facts m'am. -- Jack Webb > -Original Message- > From: John Stanton [mailto:[EMAIL PROTECTED] > Sent: Friday, November 09, 2007 5:51 PM > To: sqlite-users@sqlite.org > Subject: Re: [sqlite] Suggests for improving the SQLite website > > > James Dennett wrote: > > Joe Wilson wrote: > > > >> --- "Samuel R. Neff" <[EMAIL PROTECTED]> wrote: > >>> I hope this doesn't offend, but perhaps the best solution is to > >> outsource > >>> the website to someone or a company that specializes in > websites and > >> design > >>> (with your stated simplicity goals in mind of course). > We certainly > >>> wouldn't want a graphic designer hacking away at the > SQLite engine, > > so > >> isn't > >>> the reverse also true? > >>> > >>> Sam > >> +1 > >> > >> Also, non-technical people would be a better judge of which website > >> design is appealing. > > > > Appealing *to* non-technical people? Why would a website > on an embedded > > database wish to appeal primarily to such an audience? I'd think it > > would be best to present information in a way that appeals > to its likely > > viewers. > > > > -- James > > > > How about having adding a social networking capability so that > non-technical people will have a reason to use the website. > You cannot > expect to attract them with a frugal and highly functional embedded > database library. > > -- > --- > To unsubscribe, send email to [EMAIL PROTECTED] > -- > --- > - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
For what reason would "non-technical" types need to look on the SQLite website? :-) I kind'a like it the way it is. Plain and simple, with no over wrought graphics and other worthless fluff. It is a website for a very bare bones, plain and simple database. Those who access it are not looking for entertainment, just plain and simple results. Fred > -Original Message- > From: Joe Wilson [mailto:[EMAIL PROTECTED] > Sent: Friday, November 09, 2007 4:25 PM > To: sqlite-users@sqlite.org > Subject: RE: [sqlite] Suggests for improving the SQLite website > > > --- "Samuel R. Neff" <[EMAIL PROTECTED]> wrote: > > > > I hope this doesn't offend, but perhaps the best solution > is to outsource > > the website to someone or a company that specializes in > websites and design > > (with your stated simplicity goals in mind of course). We certainly > > wouldn't want a graphic designer hacking away at the SQLite > engine, so isn't > > the reverse also true? > > > > Sam > > +1 > > Also, non-technical people would be a better judge of which website > design is appealing. > > > __ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > > -- > --- > To unsubscribe, send email to [EMAIL PROTECTED] > -- > --- > - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
[EMAIL PROTECTED] wrote: "Mark Wyszomierski" <[EMAIL PROTECTED]> wrote: I put up 4 variations. Please, everyone, offer your opinions: (1) http://sqlite.hwaci.com/v1/ No CSS of any kind. (2) http://sqlite.hwaci.com/v2/ CSS menus with rounded corners (3) http://sqlite.hwaci.com/v3/ CSS menus with square corners (4) http://sqlite.hwaci.com/v4/ CSS font specification only (2) and (3) do not work on IE6. (1) has ugly fonts, I am told. That leaves me with (4). I suppose we could go with (4) now and change it later I like (1) the best. It is simple, uncluttered, and minimalist. I don't think the fonts are ugly at all, at least not in IE6. (What's so ugly about Times? It's the most popular font in the world.) (4) is also fine, if you insist on using CSS... - Richard Klein - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
[EMAIL PROTECTED] wrote: Joe Wilson <[EMAIL PROTECTED]> wrote: It takes time to get all popular browsers working, but it leaves a good first impression with potential users of your software. It seems like a better solution would be to do the website without any CSS and then spend the days or weeks of frustration saved working on SQLite instead. AMEN! - Richard Klein - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
[EMAIL PROTECTED] wrote: Regarding the basic "look" of the site, we were considering using a style similar to the once found at ActiveState http://www.activestate.com/ However, as we started to prototype this, we wrote down a very simple CSS/Javascript-free template and after looking at it, thought that this template might actually be better. By being CSS and Javascript-free, the new design also stays closer to the minimalist spirit of SQLite. A rough prototype of what a revised website might look like can be seen at http://sqlite.hwaci.com/ Personally, I prefer the minimalist look. And nobody is especially happy with the content of the homepage. (Suggestions for what should appear on the homepage are welcomed.) I happen to like the content of the homepage (of the *existing* site, not the new one). On the left, it shows all the info that a newbie wants to know about SQLite. On the right, it shows all the news of interest to an experienced SQLite user. It thus satisfies a wide audience at a single glance. Perfect! - Richard Klein - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
[EMAIL PROTECTED] wrote: There is a new look up on the demo site at http://sqlite.hwaci.com/ It looks good on Firefox and Safari, but IE6 renders it incorrectly. Being entirely in the unix world now, I am of a mind to ignore the IE6 problem and just let lingering IE6 users see a goofed up display. I wonder if others have differing views on this. I think it's worth the effort to make it look good under IE6. I'm a big fan of open source, and am not particularly fond of Microsoft, but IE6 is still my browser of choice. Why? Simply because I have found that it still does a better job of rendering most web content (probably because most web designers optimize for IE). - Richard Klein - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
James Dennett wrote: Joe Wilson wrote: --- "Samuel R. Neff" <[EMAIL PROTECTED]> wrote: I hope this doesn't offend, but perhaps the best solution is to outsource the website to someone or a company that specializes in websites and design (with your stated simplicity goals in mind of course). We certainly wouldn't want a graphic designer hacking away at the SQLite engine, so isn't the reverse also true? Sam +1 Also, non-technical people would be a better judge of which website design is appealing. Appealing *to* non-technical people? Why would a website on an embedded database wish to appeal primarily to such an audience? I'd think it would be best to present information in a way that appeals to its likely viewers. -- James How about having adding a social networking capability so that non-technical people will have a reason to use the website. You cannot expect to attract them with a frugal and highly functional embedded database library. - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
--- P Kishor <[EMAIL PROTECTED]> wrote: > On 11/9/07, Joe Wilson <[EMAIL PROTECTED]> wrote: > > --- James Dennett <[EMAIL PROTECTED]> wrote: > > > > > Joe Wilson wrote: > > > > Also, non-technical people would be a better judge of which website > > > > design is appealing. > > > > > > Appealing *to* non-technical people? Why would a website on an embedded > > > database wish to appeal primarily to such an audience? I'd think it > > > would be best to present information in a way that appeals to its likely > > > viewers. > > > > My wording was poor. I think a lot of programmers don't care what a software > > website looks like, as long as the code works. > > Wrong. Many, many programmers do care a lot about the looks of a > software, the code behind the software, stuff written with and for > that software, and the website of that software. While not always > co-relating, a well laid out website is also attractive, and an > attractive website is also well laid out. Software makers and > designers, particularly in the world of Macs, are very proud of how > their products look and behave, and they spend a considerable amount > of effort making them look good. Of course we all like nice looking software and websites. No one is disputing this. The question is how to achieve it. Programmers are not necessarily the best people to make aesthetic decisions. Apple has their Human Interface Group. Apple software tends to look better largely because someone has already made most of these design decisions in their famous Apple Human Interface Guidelines. There are people out there who do this sort of thing for a living, and frankly, are much better at it than most programmers. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] How to get 3.4.2 code
Mark Brown wrote: How do I know what version of the file was part of 3.4.2 and what is part of 3.5? I'm seeing version numbers of 1.171, for example. Mark, I was talking about using the sqlite download page at http://www.sqlite.org/download.html. If you right click on a link (like the one for sqlite3-3_5_2.zip under Precompiled Binaries For Windows) and select "copy the link location" (at least that's the command using Firefox), you will have a link to the specified file on the clipboard. If you then paste that file name into your browser's address bar you will have a URL like this http://www.sqlite.org/sqlite-3_5_2.zip that you can edit. If you now change the 5 in that URL to a 4 and then press enter, you will start a download of sqlite-3_4_2.zip. If you want you can simply type in the URL to start the download. All the old versions are still on the server, they just don't have clickable links on the download page, so you have to enter the filename with the desired version manually (or by editing a very similar URL using copy and paste to minimize typing errors). HTH Dennis Cote - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
--- James Dennett <[EMAIL PROTECTED]> wrote: > Joe Wilson wrote: > > Also, non-technical people would be a better judge of which website > > design is appealing. > > Appealing *to* non-technical people? Why would a website on an embedded > database wish to appeal primarily to such an audience? I'd think it > would be best to present information in a way that appeals to its likely > viewers. My wording was poor. I think a lot of programmers don't care what a software website looks like, as long as the code works. The old website was sufficient in that regard. Maybe I'm completely wrong, but I thought the point of the website redesign was to reach a new market. People who don't code, yet make development decisions for their companies. Everyone on this mailing list already is a user a SQLite. Sometimes it takes an outsider to give you an objective opinion. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] How to get 3.4.2 code
Ahh...I think I figured it out. Clicking the "Show Milestones" button was very helpful! - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] How to get 3.4.2 code
Hi Dennis- Thanks for the reply. How do I know what version of the file was part of 3.4.2 and what is part of 3.5? I'm seeing version numbers of 1.171, for example. Thanks, Mark > Mark, > > Go to the download page, copy the link for the file you want > (with the > current 3..5.2 version number), paste the link into your browser's > address bar, then edit the link and change the version number to the > version you want. All the old files are still on the server, > there just > aren't any links so you have to type them in manually. > > HTH > Dennis Cote - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
Joe Wilson wrote: > --- "Samuel R. Neff" <[EMAIL PROTECTED]> wrote: > > > > I hope this doesn't offend, but perhaps the best solution is to > outsource > > the website to someone or a company that specializes in websites and > design > > (with your stated simplicity goals in mind of course). We certainly > > wouldn't want a graphic designer hacking away at the SQLite engine, so > isn't > > the reverse also true? > > > > Sam > > +1 > > Also, non-technical people would be a better judge of which website > design is appealing. Appealing *to* non-technical people? Why would a website on an embedded database wish to appeal primarily to such an audience? I'd think it would be best to present information in a way that appeals to its likely viewers. -- James - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
--- "Samuel R. Neff" <[EMAIL PROTECTED]> wrote: > > I hope this doesn't offend, but perhaps the best solution is to outsource > the website to someone or a company that specializes in websites and design > (with your stated simplicity goals in mind of course). We certainly > wouldn't want a graphic designer hacking away at the SQLite engine, so isn't > the reverse also true? > > Sam +1 Also, non-technical people would be a better judge of which website design is appealing. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] How to get 3.4.2 code
Mark Brown wrote: Could someone please remind me how I can download SQLite code for version 3.4.2 from CVS? Was there some sort of tag made that I can use? I'm hesitant to upgrade to 3.5.x just because I'm not sure how much it has been test driven and we are close to release. Maybe I am being overly concerned, and 3.5.x is actually much more stable than 3.4.x? Mark, Go to the download page, copy the link for the file you want (with the current 3..5.2 version number), paste the link into your browser's address bar, then edit the link and change the version number to the version you want. All the old files are still on the server, there just aren't any links so you have to type them in manually. HTH Dennis Cote - To unsubscribe, send email to [EMAIL PROTECTED] -
[sqlite] Problem creating extension for use with load_extension
First off, I would like to say that although I have a lot of experience
with programming, most of it is in C#/Java and I do not have a lot of
experience with C++, although I have been working with SQLite for years.
I am attempting to create my own extension to use with SQLite but am
having problems. Using the command line interface, when I load the
extension I get the following:
SQLite version 3.5.2
Enter ".help" for instructions
sqlite> select load_extension('mydblib.dll');
SQL error: The specified procedure could not be found.
I have tried numerous things to get this to work but all have yielded
the same result. I tried modifying the SQLite source to manually add
the function I created, modifying func.c, and was successfully able to
use it in the CLI. However, I would much prefer to use an extension for
this purpose to avoid changing the source at each new release. I am
implementing an aggregate function for concatenating results in a group
by field (similar to group_concat in MySQL) and have implemented the
necessary step and finalize methods for this and passing to
sqlite3_create_function.
I am using Visual Studio 2005 to compile the project but also tried
using MinGW and had similar issues. It could very well be that I am
simply compiling the library incorrectly as I do not have a lot of
experience with this in C++, but I am unsure at this point where to
point the finger (code issue/compiling issue/etcetera). I searched the
SQLite source but cannot find anything on this error. I am running this
with Windows Vista 32-bit in a command prompt run as administrator.
Thanks for any help, let me know if I need to post a link to the
source/my project for this.
Bob Dankert
-
To unsubscribe, send email to [EMAIL PROTECTED]
-
Re: [sqlite] Read and write in SQLite
On 11/9/07, Joanne Pham <[EMAIL PROTECTED]> wrote: > I have an application to writh to database with begin transaction and end > transaction. > Before end transaction another application try to connect to the database to > read the data and in the middle of the read of the second application > the first application isues the end transaction. Because the read is still > reading the data the first application return an error message that "Couldn't > end the transaction ..." ( I don't remember the exactly error message) but > the first application couldn't "end the transaction" becuase the read of the > second application. That sounds like a message returned by a wrapper you're using. > So What you said is if the second application is connected to the database > while the first application is writing then the error message SQLITE_BUSY is > return back to application and not connect to the the database right. When the first transaction tries to COMMIT, it will get SQLITE_BUSY because the second transaction is still using the database file. The first writing transaction should simply wait a short time and try again, in a loop. When the second reading transaction is done, the first will be able to COMMIT. If you don't want concurrency like this at all, so that there can only be one transaction using the database at any time, you can use BEGIN EXCLUSIVE for all transactions. If BEGIN EXCLUSIVE returns SQLITE_BUSY, then another is using the database and this one must wait. If it succeeds, no others can use the database until this transaction is done. - To unsubscribe, send email to [EMAIL PROTECTED] -
[sqlite] How to get 3.4.2 code
Hi- Could someone please remind me how I can download SQLite code for version 3.4.2 from CVS? Was there some sort of tag made that I can use? I'm hesitant to upgrade to 3.5.x just because I'm not sure how much it has been test driven and we are close to release. Maybe I am being overly concerned, and 3.5.x is actually much more stable than 3.4.x? Thanks, Mark - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Read and write in SQLite
Hi Trevor, I have an application to writh to database with begin transaction and end transaction. Before end transaction another application try to connect to the database to read the data and in the middle of the read of the second application the first application isues the end transaction. Because the read is still reading the data the first application return an error message that "Couldn't end the transaction ..." ( I don't remember the exactly error message) but the first application couldn't "end the transaction" becuase the read of the second application. So What you said is if the second application is connected to the database while the first application is writing then the error message SQLITE_BUSY is return back to application and not connect to the the database right. Thanks, JP - Original Message From: Trevor Talbot <[EMAIL PROTECTED]> To: sqlite-users@sqlite.org Sent: Friday, November 9, 2007 1:43:26 PM Subject: Re: [sqlite] Read and write in SQLite On 11/9/07, Joanne Pham <[EMAIL PROTECTED]> wrote: > I knew that SQLite doesn't allow concurrency for both read and write at the > same time. > My application is written in C++ so Is there any way that we can check if > there is any connection to the database so the second connection needs to > wait. For example my application write to the database and other application > is reading data from database so both read and write need to check if there > is any connection to the database before making the connection to the > database. I'm not clear on what problem you're trying to solve. If SQLite cannot do something now it will return SQLITE_BUSY from most calls. What other behavior do you need? - To unsubscribe, send email to [EMAIL PROTECTED] - __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com
Re: [sqlite] Read and write in SQLite
On 11/9/07, Joanne Pham <[EMAIL PROTECTED]> wrote: > I knew that SQLite doesn't allow concurrency for both read and write at the > same time. > My application is written in C++ so Is there any way that we can check if > there is any connection to the database so the second connection needs to > wait. For example my application write to the database and other application > is reading data from database so both read and write need to check if there > is any connection to the database before making the connection to the > database. I'm not clear on what problem you're trying to solve. If SQLite cannot do something now it will return SQLITE_BUSY from most calls. What other behavior do you need? - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
On 11/9/07, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote: > I put up 4 variations. Please, everyone, offer your opinions: > >(1) http://sqlite.hwaci.com/v1/ No CSS of any kind. >(2) http://sqlite.hwaci.com/v2/ CSS menus with rounded corners >(3) http://sqlite.hwaci.com/v3/ CSS menus with square corners >(4) http://sqlite.hwaci.com/v4/ CSS font specification only > > (2) and (3) do not work on IE6. (1) has ugly fonts, I am told. > That leaves me with (4). I like (1); my browsers' default fonts suit me perfectly, TYVM. (4) is therefore ugly and harder to read for no good reason. That holds true for all 4 of my primary browsers across two different platforms, incidentally. (2) and (3) feel heavy/slow, and pulldown menus are irritating to navigate. They also do not render correctly with larger font sizes. You cannot properly account for things like mobile browsers unless you make a *lot* more effort on designs like (2) and (3). If you have to expend significant effort to handle style in even just different desktop browsers, the style is not worth pursuing. Don't waste your time; keep the site simple. Those advocating more complex designs or just installing Firefox need to remember 3 important things: 1) All the web is not viewsed on a desktop. 2) All the desktops are not 2+ GHz monsters. 3) Firefox is a slow pig of a browser. See 2. Considering SQLite is a lightweight, embedded database engine, these points are extremely relevant to your target audience. - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
I hope this doesn't offend, but perhaps the best solution is to outsource the website to someone or a company that specializes in websites and design (with your stated simplicity goals in mind of course). We certainly wouldn't want a graphic designer hacking away at the SQLite engine, so isn't the reverse also true? Sam --- We're Hiring! Seeking a passionate developer to join our team building Flex based products. Position is in the Washington D.C. metro area. If interested contact [EMAIL PROTECTED] -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Friday, November 09, 2007 1:30 PM To: sqlite-users@sqlite.org Subject: Re: [sqlite] Suggests for improving the SQLite website Joe Wilson <[EMAIL PROTECTED]> wrote: > > It takes time to get all popular browsers working, but it leaves a > good first impression with potential users of your software. > It seems like a better solution would be to do the website without any CSS and then spend the days or weeks of frustration saved working on SQLite instead. -- D. Richard Hipp <[EMAIL PROTECTED]> - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
it has been my experience that the outfits who use Harold down at the Golf Club or Snake and Lizard at the Pub to give the management their "Microsoft Only" policy are not ready to make an exception to that policy for Sqlite. Wilson, Ron wrote: That sounds great and all, but lucky you that you get to work in an office that actually lets you install firefox. Many corporations severely limit user freedom on company hardware. I'm not one of them, but I have many friends that work in these environments where only 'authorized' tools are installed on their systems for them by IT folks. Which brings us back to the question of audience. It is not just the coders that need to see the site. We have to be able to 'sell' SQLite to managers and stakeholders that wouldn't know firefox (or databases) from a hole in the ground. Often a glossy looking website (that works in most browsers) is enough to reassure them that the code base is not junk. "Hey boss - we want to use SQLite." "Sure what can you tell me about it?" "If you just install firefox, I'll show you their website." RW Ron Wilson, Senior Engineer, MPR Associates, 518.831.7546 -Original Message- From: John Stanton [mailto:[EMAIL PROTECTED] Sent: Friday, November 09, 2007 2:51 PM To: sqlite-users@sqlite.org Subject: Re: [sqlite] Suggests for improving the SQLite website To spend time working around IE deficiencies is rather futile when a simple fix exists for all users - load Firefox. It is a change which is good for them in other ways so it is a Hippocratic change - "First, do no harm". - To unsubscribe, send email to [EMAIL PROTECTED] - - To unsubscribe, send email to [EMAIL PROTECTED] - - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
Yep. The 'T' in SUPPORT disappears for me. The fonts in Linux may differ slightly from Windows. Maybe my machine does not have Verdana and is using the sans-serif backup font choice. (2) has the same issue. --- "Evans, Mark (Tandem)" <[EMAIL PROTECTED]> wrote: > I like 3 too. There's a minor glitch on Firefox - increasing font size > causes the right side of menu bar to be whited out but display when > cursor hovers. Is this the bug Joe refers to below? > > Mark > > > -Original Message- > > From: Joe Wilson [mailto:[EMAIL PROTECTED] > > Sent: Friday, November 09, 2007 1:30 PM > > To: sqlite-users@sqlite.org > > Subject: Re: [sqlite] Suggests for improving the SQLite website > > > > --- [EMAIL PROTECTED] wrote: > > > I put up 4 variations. Please, everyone, offer your opinions: > > > > > >(1) http://sqlite.hwaci.com/v1/ No CSS of any kind. > > >(2) http://sqlite.hwaci.com/v2/ CSS menus with > > rounded corners > > >(3) http://sqlite.hwaci.com/v3/ CSS menus with square corners > > >(4) http://sqlite.hwaci.com/v4/ CSS font specification only > > > > > > (2) and (3) do not work on IE6. (1) has ugly fonts, I am told. > > > That leaves me with (4). > > > > > > I suppose we could go with (4) now and change it later > > > > I prefer (3). (SUPPORT 'T' render bug in Firefox aside). > > > > (4) would also be good if you just centered the contents as > > in (3) to look better on wider resolutions. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
--- John Stanton <[EMAIL PROTECTED]> wrote: > Joe Wilson wrote: > >>(3) http://sqlite.hwaci.com/v3/ CSS menus with square corners > > > > In Firefox 2.0.0.8, press "CTRL +" a couple of times to see the render > > problem. If I press "CTRL -" it renders properly. > > > > On larger screen resoltions, sometimes the default fonts are a bit bigger > > than usual. > > > Unfortunately fonts cannot be relied upon to scale smoothly so you get > truncation and similar situations occurring at certain resolutions or > font resizing. Most popular websites will accommodate slight font size variation in their layouts. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
That sounds great and all, but lucky you that you get to work in an office that actually lets you install firefox. Many corporations severely limit user freedom on company hardware. I'm not one of them, but I have many friends that work in these environments where only 'authorized' tools are installed on their systems for them by IT folks. Which brings us back to the question of audience. It is not just the coders that need to see the site. We have to be able to 'sell' SQLite to managers and stakeholders that wouldn't know firefox (or databases) from a hole in the ground. Often a glossy looking website (that works in most browsers) is enough to reassure them that the code base is not junk. "Hey boss - we want to use SQLite." "Sure what can you tell me about it?" "If you just install firefox, I'll show you their website." RW Ron Wilson, Senior Engineer, MPR Associates, 518.831.7546 -Original Message- From: John Stanton [mailto:[EMAIL PROTECTED] Sent: Friday, November 09, 2007 2:51 PM To: sqlite-users@sqlite.org Subject: Re: [sqlite] Suggests for improving the SQLite website To spend time working around IE deficiencies is rather futile when a simple fix exists for all users - load Firefox. It is a change which is good for them in other ways so it is a Hippocratic change - "First, do no harm". - To unsubscribe, send email to [EMAIL PROTECTED] - - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
Joe Wilson wrote: (3) http://sqlite.hwaci.com/v3/ CSS menus with square corners In Firefox 2.0.0.8, press "CTRL +" a couple of times to see the render problem. If I press "CTRL -" it renders properly. On larger screen resoltions, sometimes the default fonts are a bit bigger than usual. Unfortunately fonts cannot be relied upon to scale smoothly so you get truncation and similar situations occurring at certain resolutions or font resizing. - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
I like 3 too. There's a minor glitch on Firefox - increasing font size causes the right side of menu bar to be whited out but display when cursor hovers. Is this the bug Joe refers to below? Mark > -Original Message- > From: Joe Wilson [mailto:[EMAIL PROTECTED] > Sent: Friday, November 09, 2007 1:30 PM > To: sqlite-users@sqlite.org > Subject: Re: [sqlite] Suggests for improving the SQLite website > > --- [EMAIL PROTECTED] wrote: > > I put up 4 variations. Please, everyone, offer your opinions: > > > >(1) http://sqlite.hwaci.com/v1/ No CSS of any kind. > >(2) http://sqlite.hwaci.com/v2/ CSS menus with > rounded corners > >(3) http://sqlite.hwaci.com/v3/ CSS menus with square corners > >(4) http://sqlite.hwaci.com/v4/ CSS font specification only > > > > (2) and (3) do not work on IE6. (1) has ugly fonts, I am told. > > That leaves me with (4). > > > > I suppose we could go with (4) now and change it later > > I prefer (3). (SUPPORT 'T' render bug in Firefox aside). > > (4) would also be good if you just centered the contents as > in (3) to look better on wider resolutions. > > > __ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection > around http://mail.yahoo.com > > -- > --- > To unsubscribe, send email to [EMAIL PROTECTED] > -- > --- > > - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
[EMAIL PROTECTED] wrote: bash <[EMAIL PROTECTED]> wrote: On 09/11/2007, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote: This is indeed a sad commentary on the state of the world wide web that it is now necessary to specify a font on every web page Oh well. What's about CSS? It should help in this case. I recognize that the problem is easy to fix. I am lamenting that the problem exists in the first place and that a fix is required. It used to be that you could just put up HTML and expect it to look decent. But now we have "advanced" to the point where you have to play lots of games with CSS and javascript in order to get good looks. Is this really progress? But I am off subject... There is a new look up on the demo site at http://sqlite.hwaci.com/ It looks good on Firefox and Safari, but IE6 renders it incorrectly. Being entirely in the unix world now, I am of a mind to ignore the IE6 problem and just let lingering IE6 users see a goofed up display. I wonder if others have differing views on this. -- D. Richard Hipp <[EMAIL PROTECTED]> To spend time working around IE deficiencies is rather futile when a simple fix exists for all users - load Firefox. It is a change which is good for them in other ways so it is a Hippocratic change - "First, do no harm". - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
[EMAIL PROTECTED] wrote: Joe Wilson <[EMAIL PROTECTED]> wrote: The font change to font-family: "Verdana" "sans-serif"; makes a huge difference - much more professional looking. This is indeed a sad commentary on the state of the world wide web that it is now necessary to specify a font on every web page Oh well. -- D. Richard Hipp <[EMAIL PROTECTED]> The concept of entropy. Chaos and confusion increases unless energy is expended to reverse it. The WWW is a good example of how elegant simplicity has degenerated. The applied effort produces an ever decreasing amount of result. - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
Joe Wilson wrote: > > --- James Dennett <[EMAIL PROTECTED]> wrote: > > > Joe Wilson wrote: > > > No need to say it's stable or recommended - it's assumed. Otherwise it > > > wouldn't appear on the home page. > > > > I disagree. 3.5.0 appeared, even though discussion was that it was > > relatively experimental. It's *good* to be explicit about this. > > Latest Stable Release: 2007-11-05 version 3.5.2 > Latest Dev Release:2038-01-01 version 3.9.7 > > Is more succinct and concise. Works for me. (And usually in the case of SQLite the latest dev release is also the latest stable release, and life is simple.) -- James - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
--- [EMAIL PROTECTED] wrote: > I put up 4 variations. Please, everyone, offer your opinions: > >(1) http://sqlite.hwaci.com/v1/ No CSS of any kind. >(2) http://sqlite.hwaci.com/v2/ CSS menus with rounded corners >(3) http://sqlite.hwaci.com/v3/ CSS menus with square corners >(4) http://sqlite.hwaci.com/v4/ CSS font specification only > > (2) and (3) do not work on IE6. (1) has ugly fonts, I am told. > That leaves me with (4). > > I suppose we could go with (4) now and change it later I prefer (3). (SUPPORT 'T' render bug in Firefox aside). (4) would also be good if you just centered the contents as in (3) to look better on wider resolutions. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Re: Disk caching impacts performance.
Did you run your test using ReadFile? Brandon, Nicholas (UK) wrote: I just tried (hadn't noticed that option before) to go from 2000 to 4000 and 8000, without noticing any difference. I might try next week to raise the page size to 50k and see if it makes a difference? On the presumption the Sqlite allocates new pages on the page boundary I would suggest you use multiples of the file system page size. I believe (but I could be wrong) both Linux and Windows NTFS defaults to 4096 bytes. If you chose 4,500 bytes in this scenario, for every call to retrieve one Sqlite page, it would require two calls to the OS. For whats it worth, when I did some performance testing a few years ago I recall there was no significant differenence using larger page sizes on a standard desktop machine but your mileage may vary. On a related note I do remember that when I was testing large reads (1M, 10M , 100M) using a single 'fread' call compiled in MSVC 2005 on Windows XP SP2 it had an interesting side effect in the fact that it performed the function by calling a lower level API multiple times with a size of 65,355 bytes, regardless of the original size requested in 'fread'. I believe this can be shown using the file system tools from SysInternals. Therefore I suspect there will be little-to-no benefit of page sizes greater than 64kiB on Windows XP. Nick This email and any attachments are confidential to the intended recipient and may also be privileged. If you are not the intended recipient please delete it from your system and notify the sender. You should not copy it or use it for any purpose nor disclose or distribute its contents to any other person. - To unsubscribe, send email to [EMAIL PROTECTED] - - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
--- [EMAIL PROTECTED] wrote: > I don't think the extra bandwidth is an issue. Dan points out that > if you put the CSS in a separate file, then sometimes a browser > will render the page without CSS, then when the CSS arrives a > fraction of a second later, everything shifts. That's not the case with external CSS file(s). The browser blocks until the external css file is loaded. Nothing is rendered until that time. Might you be refering to when images are loaded and cause layout to shift? If the CSS does not load due to HTTP error, then I guess it's possible that the page will be rendered without it, but I don't know for sure what happens in that case. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
Richard,
I just recently discovered that IE supports "conditionl comments", which allow
you to, among other things, load specific CSS in IE. For detail, see
http://msdn2.microsoft.com/en-us/library/ms537512.aspx
I was able to use this feature to my advantage on a project to load the main
CSS file for all browsers, and maintain a small "delta" CSS file for IE that
replaces a couple of specific classes that IE handled differently than FF.
If you'll add this as the first block in your page style:
* {
font-family: Verdana, sans-serif; /* this is the correct syntax for a CSS
font list, accordint to TopStyle */
font-size: 12px;
}
it'll set the global default font and size.
And if you change to:
.toolbar a {
color: white;
text-decoration: none;
}
IE will display the correct color for the toolbar links. Neither change
appears to affect the page display in FF.
I may find time over the weekend to see if I can fix the radiused corners, and
the broken cascading hover menus on the toolbar under IE6.
-Clark
- Original Message
From: "[EMAIL PROTECTED]" <[EMAIL PROTECTED]>
To: sqlite-users@sqlite.org
Sent: Friday, November 9, 2007 10:27:21 AM
Subject: Re: [sqlite] Suggests for improving the SQLite website
Joe Wilson <[EMAIL PROTECTED]> wrote:
>
> You chose to embed the CSS settings on each page to avoid the round
> trip to the web server. You can always put the css info in a separate
> file, and define it only once for the entire site. It should reduce
> the number of bytes sent over the wire.
>
I don't think the extra bandwidth is an issue. Dan points out that
if you put the CSS in a separate file, then sometimes a browser
will render the page without CSS, then when the CSS arrives a
fraction of a second later, everything shifts. I'd rather avoid
that.
I am now also told that web pages need to be designed for
three separate browsers: IE6, IE7, and all others. This is
madness.
--
D. Richard Hipp <[EMAIL PROTECTED]>
-
To unsubscribe, send email to [EMAIL PROTECTED]
-
-
To unsubscribe, send email to [EMAIL PROTECTED]
-
Re: [sqlite] Suggests for improving the SQLite website
[EMAIL PROTECTED] wrote: > >(1) http://sqlite.hwaci.com/v1/ No CSS of any kind. >(2) http://sqlite.hwaci.com/v2/ CSS menus with rounded corners >(3) http://sqlite.hwaci.com/v3/ CSS menus with square corners >(4) http://sqlite.hwaci.com/v4/ CSS font specification only > > (2) and (3) do not work on IE6. (1) has ugly fonts, I am told. > That leaves me with (4). > > I suppose we could go with (4) now and change it later To be honest I thought that they all looked pretty good. 1 was pretty nice and I rather liked 2 and 3. I don't know about IE6, but they looked fine in IE7. Clay -- Lazarus Registration http://www.lazarusid.com/registration.shtml - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
--- James Dennett <[EMAIL PROTECTED]> wrote: > Joe Wilson wrote: > > No need to say it's stable or recommended - it's assumed. Otherwise it > > wouldn't appear on the home page. > > I disagree. 3.5.0 appeared, even though discussion was that it was > relatively experimental. It's *good* to be explicit about this. Latest Stable Release: 2007-11-05 version 3.5.2 Latest Dev Release:2038-01-01 version 3.9.7 Is more succinct and concise. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
I think 4 looks great, Mark On Nov 9, 2007 1:45 PM, <[EMAIL PROTECTED]> wrote: > "Mark Wyszomierski" <[EMAIL PROTECTED]> wrote: > > Not a terribly useful comment but was just glancing through the new > > look and noticed a typo: > > > > http://sqlite.hwaci.com/about.html > > > > "We believe that General Electric uses SQLite in some product or > > another because they twice wrote the to SQLite developers ".. > > > > "wrote the to " > > > > Thanks, Mark. I am going to go through and clean all that up. > I'm focused on the layout right now, though. > > I put up 4 variations. Please, everyone, offer your opinions: > > (1) http://sqlite.hwaci.com/v1/ No CSS of any kind. > (2) http://sqlite.hwaci.com/v2/ CSS menus with rounded corners > (3) http://sqlite.hwaci.com/v3/ CSS menus with square corners > (4) http://sqlite.hwaci.com/v4/ CSS font specification only > > (2) and (3) do not work on IE6. (1) has ugly fonts, I am told. > That leaves me with (4). > > I suppose we could go with (4) now and change it later > > -- > D. Richard Hipp <[EMAIL PROTECTED]> > > > > - > To unsubscribe, send email to [EMAIL PROTECTED] > - > > - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
I'd echo the suggestion for making the directions for import/export more prominent (which may mean simply linked from multiple places). Perhaps I'd also suggest linked to the command-line tool information from several places, maybe including the SYNTAX area. Newcomers seem to be able to miss it often. Great site and product, btw. - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
I vote for (4). --- Marco Bambini http://www.sqlabs.net http://www.sqlabs.net/blog/ http://www.sqlabs.net/realsqlserver/ On Nov 9, 2007, at 7:45 PM, [EMAIL PROTECTED] wrote: "Mark Wyszomierski" <[EMAIL PROTECTED]> wrote: Not a terribly useful comment but was just glancing through the new look and noticed a typo: http://sqlite.hwaci.com/about.html "We believe that General Electric uses SQLite in some product or another because they twice wrote the to SQLite developers ".. "wrote the to " Thanks, Mark. I am going to go through and clean all that up. I'm focused on the layout right now, though. I put up 4 variations. Please, everyone, offer your opinions: (1) http://sqlite.hwaci.com/v1/ No CSS of any kind. (2) http://sqlite.hwaci.com/v2/ CSS menus with rounded corners (3) http://sqlite.hwaci.com/v3/ CSS menus with square corners (4) http://sqlite.hwaci.com/v4/ CSS font specification only (2) and (3) do not work on IE6. (1) has ugly fonts, I am told. That leaves me with (4). I suppose we could go with (4) now and change it later -- D. Richard Hipp <[EMAIL PROTECTED]> -- --- To unsubscribe, send email to [EMAIL PROTECTED] -- --- - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
(4) is my choice. I'm not fond of the drop down menu scripts on webpages anyway... but that's just me. RW Ron Wilson, Senior Engineer, MPR Associates, 518.831.7546 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Friday, November 09, 2007 1:45 PM To: sqlite-users@sqlite.org Subject: Re: [sqlite] Suggests for improving the SQLite website "Mark Wyszomierski" <[EMAIL PROTECTED]> wrote: > Not a terribly useful comment but was just glancing through the new > look and noticed a typo: > > http://sqlite.hwaci.com/about.html > > "We believe that General Electric uses SQLite in some product or > another because they twice wrote the to SQLite developers ".. > > "wrote the to " > Thanks, Mark. I am going to go through and clean all that up. I'm focused on the layout right now, though. I put up 4 variations. Please, everyone, offer your opinions: (1) http://sqlite.hwaci.com/v1/ No CSS of any kind. (2) http://sqlite.hwaci.com/v2/ CSS menus with rounded corners (3) http://sqlite.hwaci.com/v3/ CSS menus with square corners (4) http://sqlite.hwaci.com/v4/ CSS font specification only (2) and (3) do not work on IE6. (1) has ugly fonts, I am told. That leaves me with (4). I suppose we could go with (4) now and change it later -- D. Richard Hipp <[EMAIL PROTECTED]> - To unsubscribe, send email to [EMAIL PROTECTED] - - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
Another solution is to design your css for standard browser and then just create a iefixes.css file to load only in IE that contains the various fixes for that browser. The trick is to add that lines in the head section: --- Marco Bambini http://www.sqlabs.net http://www.sqlabs.net/blog/ http://www.sqlabs.net/realsqlserver/ On Nov 9, 2007, at 7:29 PM, [EMAIL PROTECTED] wrote: Joe Wilson <[EMAIL PROTECTED]> wrote: It takes time to get all popular browsers working, but it leaves a good first impression with potential users of your software. It seems like a better solution would be to do the website without any CSS and then spend the days or weeks of frustration saved working on SQLite instead. -- D. Richard Hipp <[EMAIL PROTECTED]> -- --- To unsubscribe, send email to [EMAIL PROTECTED] -- --- - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
That looks fantastic (in firefox). Ron Wilson, Senior Engineer, MPR Associates, 518.831.7546 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Friday, November 09, 2007 1:11 PM To: sqlite-users@sqlite.org Subject: Re: [sqlite] Suggests for improving the SQLite website [...] There is a new look up on the demo site at http://sqlite.hwaci.com/ It looks good on Firefox and Safari, but IE6 renders it incorrectly. Being entirely in the unix world now, I am of a mind to ignore the IE6 problem and just let lingering IE6 users see a goofed up display. I wonder if others have differing views on this. -- D. Richard Hipp <[EMAIL PROTECTED]> - To unsubscribe, send email to [EMAIL PROTECTED] - - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
> I don't think the extra bandwidth is an issue. Dan points out that > if you put the CSS in a separate file, then sometimes a browser > will render the page without CSS, then when the CSS arrives a > fraction of a second later, everything shifts. I'd rather avoid > that. > > I am now also told that web pages need to be designed for > three separate browsers: IE6, IE7, and all others. This is > madness. Richard, talk to SteveL. IIRC he did the CSS for the Tcler's Wiki recently, or at least knows who did ... Might have been Jos DeCoster too. And that CSS has IIRC a lot of the tricks for nice display on a variety of browsers. +1 on the madness. -- Andreas Kupries <[EMAIL PROTECTED]> Developer @ http://www.ActiveState.com Tel: +1 778-786-1122 - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
"Mark Wyszomierski" <[EMAIL PROTECTED]> wrote: > Not a terribly useful comment but was just glancing through the new > look and noticed a typo: > > http://sqlite.hwaci.com/about.html > > "We believe that General Electric uses SQLite in some product or > another because they twice wrote the to SQLite developers ".. > > "wrote the to " > Thanks, Mark. I am going to go through and clean all that up. I'm focused on the layout right now, though. I put up 4 variations. Please, everyone, offer your opinions: (1) http://sqlite.hwaci.com/v1/ No CSS of any kind. (2) http://sqlite.hwaci.com/v2/ CSS menus with rounded corners (3) http://sqlite.hwaci.com/v3/ CSS menus with square corners (4) http://sqlite.hwaci.com/v4/ CSS font specification only (2) and (3) do not work on IE6. (1) has ugly fonts, I am told. That leaves me with (4). I suppose we could go with (4) now and change it later -- D. Richard Hipp <[EMAIL PROTECTED]> - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
drh wrote: > There is a new look up on the demo site at > >http://sqlite.hwaci.com/ It does look much "prettier" than the current live site. I note the common "affect"/"effect" typo in the sentence "There are no known issues effecting database integrity or correctness.", where this should be "affecting" -- SQLite does an excellent job of effecting database integrity and correctness :) -- James - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
Joe Wilson <[EMAIL PROTECTED]> wrote: > > It takes time to get all popular browsers working, but it leaves a > good first impression with potential users of your software. > It seems like a better solution would be to do the website without any CSS and then spend the days or weeks of frustration saved working on SQLite instead. -- D. Richard Hipp <[EMAIL PROTECTED]> - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
Not a terribly useful comment but was just glancing through the new look and noticed a typo: http://sqlite.hwaci.com/about.html "We believe that General Electric uses SQLite in some product or another because they twice wrote the to SQLite developers ".. "wrote the to " On Nov 9, 2007 1:22 PM, Joe Wilson <[EMAIL PROTECTED]> wrote: > --- [EMAIL PROTECTED] wrote: > > There is a new look up on the demo site at > > > >http://sqlite.hwaci.com/ > > > > It looks good on Firefox and Safari, but IE6 renders > > The 'T' in 'SUPPORT' in the horizontal toolbar is cut off in my Linux > Firefox 2.0.0.8 browser. I have a screen resolution of 1600x1200. > > It takes time to get all popular browsers working, but it leaves a > good first impression with potential users of your software. > > > __ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > > - > To unsubscribe, send email to [EMAIL PROTECTED] > - > > - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
Joe Wilson wrote: > No need to say it's stable or recommended - it's assumed. Otherwise it > wouldn't appear on the home page. I disagree. 3.5.0 appeared, even though discussion was that it was relatively experimental. It's *good* to be explicit about this. -- James - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
Joe Wilson <[EMAIL PROTECTED]> wrote: > > You chose to embed the CSS settings on each page to avoid the round > trip to the web server. You can always put the css info in a separate > file, and define it only once for the entire site. It should reduce > the number of bytes sent over the wire. > I don't think the extra bandwidth is an issue. Dan points out that if you put the CSS in a separate file, then sometimes a browser will render the page without CSS, then when the CSS arrives a fraction of a second later, everything shifts. I'd rather avoid that. I am now also told that web pages need to be designed for three separate browsers: IE6, IE7, and all others. This is madness. -- D. Richard Hipp <[EMAIL PROTECTED]> - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
> There is a new look up on the demo site at > >http://sqlite.hwaci.com/ > > It looks good on Firefox and Safari, but IE6 renders > it incorrectly. Being entirely in the unix world now, > I am of a mind to ignore the IE6 problem and just let > lingering IE6 users see a goofed up display. I wonder > if others have differing views on this. It renders pretty much fine with IE7. The only major difference between IE7 and Firefox is that with IE7 the nav bar links show up as blue/purple instead of always white. -Jeff - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
--- [EMAIL PROTECTED] wrote: > There is a new look up on the demo site at > >http://sqlite.hwaci.com/ > > It looks good on Firefox and Safari, but IE6 renders The 'T' in 'SUPPORT' in the horizontal toolbar is cut off in my Linux Firefox 2.0.0.8 browser. I have a screen resolution of 1600x1200. It takes time to get all popular browsers working, but it leaves a good first impression with potential users of your software. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
--- [EMAIL PROTECTED] wrote: > Joe Wilson <[EMAIL PROTECTED]> wrote: > > The font change to > > > > font-family: "Verdana" "sans-serif"; > > > > makes a huge difference - much more professional looking. > > > > This is indeed a sad commentary on the state of the > world wide web that it is now necessary to specify > a font on every web page Oh well. I think sqlite.org was the last site in the world that did not specify a font. But now I can't render the new home page on my original Tim Berners-Lee web browser on my NeXT cube! (just kidding) You chose to embed the CSS settings on each page to avoid the round trip to the web server. You can always put the css info in a separate file, and define it only once for the entire site. It should reduce the number of bytes sent over the wire. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
bash <[EMAIL PROTECTED]> wrote: > On 09/11/2007, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote: > > This is indeed a sad commentary on the state of the > > world wide web that it is now necessary to specify > > a font on every web page Oh well. > > > What's about CSS? It should help in this case. > I recognize that the problem is easy to fix. I am lamenting that the problem exists in the first place and that a fix is required. It used to be that you could just put up HTML and expect it to look decent. But now we have "advanced" to the point where you have to play lots of games with CSS and javascript in order to get good looks. Is this really progress? But I am off subject... There is a new look up on the demo site at http://sqlite.hwaci.com/ It looks good on Firefox and Safari, but IE6 renders it incorrectly. Being entirely in the unix world now, I am of a mind to ignore the IE6 problem and just let lingering IE6 users see a goofed up display. I wonder if others have differing views on this. -- D. Richard Hipp <[EMAIL PROTECTED]> - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
Just put the font-family in your .css file and all the pages that refer to your .css will conform. You already said that the demo site was .css driven... RW Ron Wilson, Senior Engineer, MPR Associates, 518.831.7546 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Friday, November 09, 2007 12:47 PM To: sqlite-users@sqlite.org Subject: Re: [sqlite] Suggests for improving the SQLite website Joe Wilson <[EMAIL PROTECTED]> wrote: > The font change to > > font-family: "Verdana" "sans-serif"; > > makes a huge difference - much more professional looking. > This is indeed a sad commentary on the state of the world wide web that it is now necessary to specify a font on every web page Oh well. -- D. Richard Hipp <[EMAIL PROTECTED]> - To unsubscribe, send email to [EMAIL PROTECTED] - - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
--- bash <[EMAIL PROTECTED]> wrote: > I am owner of web site with ~ 15k uniq visitors/day and I can say this > is not really true. > For example this is statistics from google analytics: > 1.1024x76842.51% > 2.1280x1024 27.73% > 3.1280x80010.43% > 4.1152x8645.33% > 5.1440x9003.10% For some reason many PCs ship Windows with much lower resolutions than their graphics cards and monitors can support - even laptops. It's a 2 second change for someone technical, but a lot of users aren't even aware of the issue and leave the default setting. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
On 09/11/2007, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote: > This is indeed a sad commentary on the state of the > world wide web that it is now necessary to specify > a font on every web page Oh well. What's about CSS? It should help in this case. -- Biomechanica Artificial Sabotage Humanoid - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
On 09/11/2007, A.J.Millan <[EMAIL PROTECTED]> wrote: > > > Regarding the basic "look" of the site, we were considering > > using a style similar to the once found at ActiveState > > > > http://www.activestate.com/ > > > > However the tendency in computers screen is wider than until now. Today the > standard is about 1440 pixels x 900, so a unique horizontal arrange y a > waste of space. The newspapers designers tend to be specialist in this, > and they use several columns to arrange the information. Obviously this is > not a newspaper, but IMHO tree columns would be good, at least in the main > page. Two for the rest. > > Good luck! I am owner of web site with ~ 15k uniq visitors/day and I can say this is not really true. For example this is statistics from google analytics: 1. 1024x76842.51% 2. 1280x1024 27.73% 3. 1280x80010.43% 4. 1152x8645.33% 5. 1440x9003.10% -- Biomechanica Artificial Sabotage Humanoid - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Re: Disk caching impacts performance.
Hello Julien, JR> JR> I can give you a few timings I have to give you an idea: JR> JR> Initialisation: connects to DB, pro-compile some queries, load JR> structured data (SELECT * FROM Objects; i.e. sequentially, no blobs) JR> Load Blobs: load 1/4 of all Blobs (each ~23k of size, I just checked), JR> "randomly" JR> JR> -Uncached JR> -14s to initialise JR> -31s to load blobs JR> JR> JR> -Uncached, but VACUUMed JR> -3s to initialise JR> -16s to load blobs JR> JR> -Pre cached, no VACUUM JR> -3s to cache and initialise JR> -1s to load blobs JR> JR> Notes: JR> - VACUUM took (as expected) a very long time to do it's task (few JR> minutes). JR> - This test is for a ~100MB file. From other tests I made it looks like JR> the timings depend linearly on size of file, and number of blobs loaded JR> JR> Pre-caching is clearly a winner here. This agrees with my own experiments. Precaching either by sequential reading or creating a file mapping and reading the data before you open the DB gives a significant boost in performance when the DB is opened the first time. C - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Re: Disk caching impacts performance.
> > I just tried (hadn't noticed that option before) to go from > 2000 to 4000 and 8000, without noticing any difference. I > might try next week to raise the page size to 50k and see if > it makes a difference? > On the presumption the Sqlite allocates new pages on the page boundary I would suggest you use multiples of the file system page size. I believe (but I could be wrong) both Linux and Windows NTFS defaults to 4096 bytes. If you chose 4,500 bytes in this scenario, for every call to retrieve one Sqlite page, it would require two calls to the OS. For whats it worth, when I did some performance testing a few years ago I recall there was no significant differenence using larger page sizes on a standard desktop machine but your mileage may vary. On a related note I do remember that when I was testing large reads (1M, 10M , 100M) using a single 'fread' call compiled in MSVC 2005 on Windows XP SP2 it had an interesting side effect in the fact that it performed the function by calling a lower level API multiple times with a size of 65,355 bytes, regardless of the original size requested in 'fread'. I believe this can be shown using the file system tools from SysInternals. Therefore I suspect there will be little-to-no benefit of page sizes greater than 64kiB on Windows XP. Nick This email and any attachments are confidential to the intended recipient and may also be privileged. If you are not the intended recipient please delete it from your system and notify the sender. You should not copy it or use it for any purpose nor disclose or distribute its contents to any other person. - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
Regarding the basic "look" of the site, we were considering using a style similar to the once found at ActiveState http://www.activestate.com/ However the tendency in computers screen is wider than until now. Today the standard is about 1440 pixels x 900, so a unique horizontal arrange y a waste of space. The newspapers designers tend to be specialist in this, and they use several columns to arrange the information. Obviously this is not a newspaper, but IMHO tree columns would be good, at least in the main page. Two for the rest. Good luck! - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
The font change to font-family: "Verdana" "sans-serif"; makes a huge difference - much more professional looking. --- [EMAIL PROTECTED] wrote: > A rough prototype of what a revised website might look like > can be seen at > > http://sqlite.hwaci.com/ __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Re: Disk caching impacts performance.
> > One more thing, did raising the limit on the number of pages SQLITE > > can cache internally have any effect? > > I just tried (hadn't noticed that option before) to go from 2000 to 4000 > and 8000, without noticing any difference. I might try next week to > raise the page size to 50k and see if it makes a difference? On an OS with file caching, this sqlite page cache setting will benefit write transactions more than reads. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe, send email to [EMAIL PROTECTED] -
[sqlite] Soft Heap Limit
Hi- I'm currently using SQLite 3.4.1 and have been reading with interest the large database thread. I learned about the soft heap limit feature and was considering using it. While reading about the bug fixes for 3.4.2 and soft heap limit, I came across a sample of setting the soft heap limit to 5000 bytes, in which Ticket 2565 (http://www.sqlite.org/cvstrac/tktview?tn=2565) indicates is a low value. I'm wondering how much memory SQLite typically uses, especially if using the default disabled soft heap limit. The reason I wonder is that we have a single thread process all of our SQLite calls, and the thread has a stack size of 16K. I'm wondering if we're going to be running into lots of memory problems without the soft heap limit in place, and if I chose to set the soft limit to, say, 2K, is performance going to suffer greatly. Just looking for some tuning adivce. Thanks, Mark - To unsubscribe, send email to [EMAIL PROTECTED] -
[sqlite] Re: Disk caching impacts performance.
Mark Spiegel wrote: > [...] Thanks for the explanation. Yes, I'll have to make sure not to use this technique when the file is too large. But I think there is definitively an improvement for us to pull data in cache whenever possible. > Given that you can't write your own VFS, there is not much to suggest, > but one question to ask. Is the sum of the time for the pre-read you > perform and the subsequent database operation(s) smaller than doing the > database operation(s) without the pre-read? I see that in the 3.5.x > source Dr. Hipp gives the file system the proper random access hint to > the file system when opening databases. This is just a hint to the > cache manager and it is not obligated to honor it, but it will > effectively shut down most read ahead and large block reads which is > what you are getting when you sequentially pre-read. I can give you a few timings I have to give you an idea: Initialisation: connects to DB, pro-compile some queries, load structured data (SELECT * FROM Objects; i.e. sequentially, no blobs) Load Blobs: load 1/4 of all Blobs (each ~23k of size, I just checked), "randomly" -Uncached -14s to initialise -31s to load blobs -Uncached, but VACUUMed -3s to initialise -16s to load blobs -Pre cached, no VACUUM -3s to cache and initialise -1s to load blobs Notes: - VACUUM took (as expected) a very long time to do it's task (few minutes). - This test is for a ~100MB file. From other tests I made it looks like the timings depend linearly on size of file, and number of blobs loaded Pre-caching is clearly a winner here. > One more thing, did raising the limit on the number of pages SQLITE can > cache internally have any effect? I just tried (hadn't noticed that option before) to go from 2000 to 4000 and 8000, without noticing any difference. I might try next week to raise the page size to 50k and see if it makes a difference?
RE: [sqlite] Suggests for improving the SQLite website
> > Please continue to provide feedback. > Assuming the build process is fairly automated and not too onerous to implement I would like to see 'nightlys/weeklys' source and precompiled binaries of SQLite. I would imagine like me, many of us are behind company firewalls with no facility for using cvs externally. The thought of downloading every file using http://www.sqlite.org/cvstrac/dir?d=sqlite is not particularly pleasing ;) Admittedly not a presentation comment but rather an improvement to what the website offers. This email and any attachments are confidential to the intended recipient and may also be privileged. If you are not the intended recipient please delete it from your system and notify the sender. You should not copy it or use it for any purpose nor disclose or distribute its contents to any other person. - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
--- [EMAIL PROTECTED] wrote: > A rough prototype of what a revised website might look like > can be seen at > > http://sqlite.hwaci.com/ Instead of: Current Status As of 2007-11-05 20:49:21 UTC, version 3.5.2 of SQLite is stable. There are no known issues effecting database integrity or correctness. Version 3.5.2 is recommended for all users. I'd recommend: Latest Release: 2007-11-05 version 3.5.2 No need to say it's stable or recommended - it's assumed. Otherwise it wouldn't appear on the home page. Or perhaps you could write: Latest Stable Release: 2007-11-05 version 3.5.2 Latest Dev Release:2038-01-01 version 3.9.7 with an appropriate link to download the specific versions. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
On Nov 9, 2007 3:11 PM, <[EMAIL PROTECTED]> wrote: > http://sqlite.hwaci.com/ i like it! i like it so much, in fact, that i'll probably buy the software! ;) > happy with the content of the homepage. (Suggestions for > what should appear on the homepage are welcomed.) a) What is sqlite? (A sentence or three, but not more.) b) License? c) Brief overview of where/why one might deploy sqlite. d) Any important links which don't quite belong (or fit) in the menu bar. Another thing i'd like to see on the front page: the "blurb" on the right-hand side of the current site, listing the latest couple of news items, is always what i look at first when i visit sqlite.org. :) -- - stephan beal http://wanderinghorse.net/home/stephan/ - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Disk caching impacts performance.
Julien said in the original post: > I've been doing some test with a ~100 MB database on a real fast physical disk it would take < 1 sec to read the entire database. On a slow drive like on my laptop, it takes 3 sec, and on my external USB drive it takes 5 sec. > Now the first time I run my application, it takes quite a long time > (30s) to load the blobs. This means we are at a transfer speed of 3Mb/s. This indicates that some seeks are involved that slow down the process. The seek time of a flash drive is a at least 4 orders of magnitude faster than the best mechanical disk. If you have a really fast mechanical disk with a seek time of 1 ms, this means you can do 1000 seeks per second. Now, let's assume we have a flash drive with 30Mb/s transfer rate and <1 ns seek time, the operation would take 3 seconds. > I tried to VACUUM the DB, it only reduced the timings to 16s This indicates that there is still quite some time lost with seeks. I think there is a general misconception on what the performance problems of physical disk drives are. Transfer speed is often not the problem, but seeks are. Databases like sqlite are optimized under the assumption that it is much better to do a few table lookups to avoid disk IO. With a sqlite database, there is no way to read the entire database in physical order. You always reed the table in ROW_ID order (if you are lucky you might have filled the database in key order or you have vacuumed it). However, if you have some more complex queries, the database engine happily seeks around on the disk. Some simple math: let's assume we have a 100mb database, and a drive with 30mb transfer rate and 10ms seek. Then (on a cold database) reading the entire database is much faster than 1000 seeks. However, once the data is in (cache) memory the rules change. It makes sense to do some index lookups to access the data. So, ideally, the operating system disk cache should be disabled for the database files and the database engine should have full control over the file and the disk cache. The database engine would need some knowledge about the attached disk (it's seek time and transfer rate) to make good decisions. Then the database engine could really optimize the database access depending on where the data is: in cache memory or on the physical disk. And, yes, write speed is a problem with solid state drives Michael John Stanton wrote: Michael Scharf wrote: [EMAIL PROTECTED] wrote: "Trevor Talbot" <[EMAIL PROTECTED]> wrote: Beyond that, I'm not aware of anything that would help. All good advice. But you left off the obvious: Get a faster disk drive. ;-) ...which does not really help unless you buy a very expensive flash disk drive. How much faster is a *really* fast spinning disk? Trevo, have you tried to put your database on a (fast!) USB stick. It should be much faster in 'seeking' but is slower in the data transfer. This would give some indication if the access is limited by seek or the disk reading speed. Michael A USB flash drive is not particularly fast due to the limited write speed of flash memory and buss speed. A fast disk spins at 15,000 rpm, double the speed of the higher end 7,500 rpm disks and almost 3 times the speed of the regular 5,400 rpm devices. If you want to simulate a disk with no latency set up a RAM drive. There is a physical constraint here. If you want to verify that your data is safely written to non-volatile storage you have to live with the latency. If that is unimportant to you you can relax the ACID requirements and get faster writes, but when you do that there is no crying over lost data after a crash. - To unsubscribe, send email to [EMAIL PROTECTED] - -- http://MichaelScharf.blogspot.com/ - To unsubscribe, send email to [EMAIL PROTECTED] -
[sqlite] Mingw32ce
I am looking for anybody who has managed to compile sqlite3 and build up any kind of application using arm-wince-mingw32ce tool. -- View this message in context: http://www.nabble.com/Mingw32ce-tf4778576.html#a13669983 Sent from the SQLite mailing list archive at Nabble.com. - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
[EMAIL PROTECTED] wrote: Thanks, everybody, for the excellent feedback and suggestions for revising the SQLite website. Please keep the comments coming. Regarding the basic "look" of the site, we were considering using a style similar to the once found at ActiveState http://www.activestate.com/ However, as we started to prototype this, we wrote down a very simple CSS/Javascript-free template and after looking at it, thought that this template might actually be better. By being CSS and Javascript-free, the new design also stays closer to the minimalist spirit of SQLite. A rough prototype of what a revised website might look like can be seen at http://sqlite.hwaci.com/ Please note that there are a lot of non-working links in this demo - it is just concept demo. And nobody is especially happy with the content of the homepage. (Suggestions for what should appear on the homepage are welcomed.) We are also working on a more elaborate concept that involves lots of CSS and javascript, pulldown menus, graphics, after the style of http://www.activestate.com/. Depending on how it looks in the end, we might or might not put the second up for review later today or Monday. Please continue to provide feedback. -- D. Richard Hipp <[EMAIL PROTECTED]> Complex web sites look stylish but quickly become tedious when used frequently. An elegant design approach is "Adaptive Input", where the user can select ever more input-driven interfaces to speed access and reward familiarity. I find that to use the Firefox Firebug debugger and look at the files loaded from a website and the time taken is very significant. The elaborate web page which loads 300 files is unmasked for what it is, a time and resource squanderer. The Activestate web page loads about 15 files and on my desktop loads and renders in 950mS. The current Sqlite.org page loads two files and completes in 500Ms, half the time of Activestate. - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] sqlite lock error
Richard, thanks a lot! I was completely sure I use the same connection, but indeed ... -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Friday, November 09, 2007 6:06 PM To: sqlite-users@sqlite.org Subject: Re: [sqlite] sqlite lock error "Maxim V. Shiyanovsky" <[EMAIL PROTECTED]> wrote: > > I start transaction and delete most of records > > After this sqlite fails on "select from sqlite_master" (or any other query) > because it extend lock levelš to exclusive lock when deleting records. > > It reset exclusive lock on transaction commit only. When one SQLite database connection is writing to the database, other database connections might be able to read, or they might be locked out completely. They get locked out complete when the write transaction is large and the cache must spill to disk. If you want to be able to read and write within a transaction, use the same database connection. -- D. Richard Hipp <[EMAIL PROTECTED]> - To unsubscribe, send email to [EMAIL PROTECTED] - - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
[EMAIL PROTECTED] wrote: A rough prototype of what a revised website might look like can be seen at http://sqlite.hwaci.com/ Short, simple, and sweet. I like it. My only specific comment was going to be a request to make the page for datatypes easy to find. Until I made it a book mark for myself, I would look at the FAQ's, search the wiki (perhaps missing it both places), and finally find an email giving the URL. If it is already easy for everyone else to find, no need to change the web page--just ban retards like me from using the product. SQLite rocks! Gerry - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] sqlite lock error
"Maxim V. Shiyanovsky" <[EMAIL PROTECTED]> wrote: > > I start transaction and delete most of records > > After this sqlite fails on "select from sqlite_master" (or any other query) > because it extend lock level to exclusive lock when deleting records. > > It reset exclusive lock on transaction commit only. When one SQLite database connection is writing to the database, other database connections might be able to read, or they might be locked out completely. They get locked out complete when the write transaction is large and the cache must spill to disk. If you want to be able to read and write within a transaction, use the same database connection. -- D. Richard Hipp <[EMAIL PROTECTED]> - To unsubscribe, send email to [EMAIL PROTECTED] -
RE: [sqlite] Suggests for improving the SQLite website
One thing I really like about the current home page is the listing of the past 4-5 versions, the date when they were released and what changed. It is so easy to see what has changed since the version that I happen to be on. That may not need to be on the front page necessarily (although I like it), but please do keep the concise summary somewhere. > -Original Message- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > Sent: Friday, November 09, 2007 8:11 AM > To: sqlite-users@sqlite.org > Subject: Re: [sqlite] Suggests for improving the SQLite website > > Thanks, everybody, for the excellent feedback and suggestions > for revising the SQLite website. Please keep the comments > coming. > > Regarding the basic "look" of the site, we were considering > using a style similar to the once found at ActiveState > >http://www.activestate.com/ > > However, as we started to prototype this, we wrote down a > very simple CSS/Javascript-free template and after looking > at it, thought that this template might actually be better. > By being CSS and Javascript-free, the new design also stays > closer to the minimalist spirit of SQLite. > > A rough prototype of what a revised website might look like > can be seen at > > http://sqlite.hwaci.com/ > > Please note that there are a lot of non-working links in this > demo - it is just concept demo. And nobody is especially > happy with the content of the homepage. (Suggestions for > what should appear on the homepage are welcomed.) > > We are also working on a more elaborate concept that involves > lots of CSS and javascript, pulldown menus, graphics, after > the style of http://www.activestate.com/. Depending on how > it looks in the end, we might or might not put the second > up for review later today or Monday. > > Please continue to provide feedback. > > -- > D. Richard Hipp <[EMAIL PROTECTED]> > > > - > To unsubscribe, send email to [EMAIL PROTECTED] > - - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Problems after upgrade
[EMAIL PROTECTED] writes: >> Profiling revealed that version 3.4.2 seems to have some kind of problem >> with the DB if created with version 3.3.8 of the library. These are the >> first lines of the profiled data on an (despite sqlite) basically idle >> system: >> >> samples %app name symbol name >> 574 7.1616 libsqlite3.so.0.8.6 sqlite3VdbeExec >> 485 6.0512 libsqlite3.so.0.8.6 __i686.get_pc_thunk.bx >> 315 3.9301 libsqlite3.so.0.8.6 sqlite3VdbeSerialType >> 265 3.3063 libsqlite3.so.0.8.6 sqlite3VdbeSerialPut > The database file format has not changed. You are the first > person to report any similar problems. Are you *certain* that > is the database upgrade that is causing your problem? Well, after profiling I stopped my application, moved the DB away and restarted. Then the problem went away. Also, after upgrading Sqlite my application was restarted, so I do not any reason to fail. Hmm... Sqlite 3.3.8 was the version as it is available on SuSE SLES10, while the updated version is the unpatched 3.4.2. Maybe I should go check the SuSE patches... /holger - To unsubscribe, send email to [EMAIL PROTECTED] -
[sqlite] sqlite lock error
Does anybody here knows sqlite internals (especially database file locking algorithm)? Or am I only one who meet this bug? May be someone could persuade me that it's not a bug? Concerning the problem - I have some table (not so big at my point of view but with 6 records). I start transaction and delete most of records After this sqlite fails on "select from sqlite_master" (or any other query) because it extend lock level to exclusive lock when deleting records. It reset exclusive lock on transaction commit only. Locking algorithm is different when table holds a few records less I've already posted sqlite trace here. May be there is known limit of how many records one could delete from table or update in single transaction? It is significant that I have single tread plain application.
Re: [sqlite] Suggests for improving the SQLite website
Thanks, everybody, for the excellent feedback and suggestions for revising the SQLite website. Please keep the comments coming. Regarding the basic "look" of the site, we were considering using a style similar to the once found at ActiveState http://www.activestate.com/ However, as we started to prototype this, we wrote down a very simple CSS/Javascript-free template and after looking at it, thought that this template might actually be better. By being CSS and Javascript-free, the new design also stays closer to the minimalist spirit of SQLite. A rough prototype of what a revised website might look like can be seen at http://sqlite.hwaci.com/ Please note that there are a lot of non-working links in this demo - it is just concept demo. And nobody is especially happy with the content of the homepage. (Suggestions for what should appear on the homepage are welcomed.) We are also working on a more elaborate concept that involves lots of CSS and javascript, pulldown menus, graphics, after the style of http://www.activestate.com/. Depending on how it looks in the end, we might or might not put the second up for review later today or Monday. Please continue to provide feedback. -- D. Richard Hipp <[EMAIL PROTECTED]> - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Problems after upgrade
Holger Eitzenberger <[EMAIL PROTECTED]> wrote: > Hi, > > I was using Sqlite library 3.3.8 before upgrading it to version 3.4.2 on > Linux. > > Right after upgrade I noticed that the CPU load of my application, which > is shared linked to the library, increased considerably. > > Profiling revealed that version 3.4.2 seems to have some kind of problem > with the DB if created with version 3.3.8 of the library. These are the > first lines of the profiled data on an (despite sqlite) basically idle > system: > > samples %app name symbol name > 574 7.1616 libsqlite3.so.0.8.6 sqlite3VdbeExec > 485 6.0512 libsqlite3.so.0.8.6 __i686.get_pc_thunk.bx > 315 3.9301 libsqlite3.so.0.8.6 sqlite3VdbeSerialType > 265 3.3063 libsqlite3.so.0.8.6 sqlite3VdbeSerialPut > 263 3.2813 libsqlite3.so.0.8.6 sqlite3PutVarint > 228 2.8447 audld.plx(no symbols) > 227 2.8322 libc-2.4.so _int_malloc > 212 2.6450 libfreetype.so.6.3.8 (no symbols) > 200 2.4953 vmlinux-2.6.16.43-54-default default_idle > 182 2.2707 libcsscan.so convert_line > 160 1.9963 libsqlite3.so.0.8.6 sqlite3PagerAcquire > 149 1.8590 confd.plx(no symbols) > ... > > Now, despite removing the old lib, is there some chance to keep the old > DB and kind of "repair" it? > > Also, is there some kind of policy which tells me if the format of > the DB changed? > The database file format has not changed. You are the first person to report any similar problems. Are you *certain* that is the database upgrade that is causing your problem? -- D. Richard Hipp <[EMAIL PROTECTED]> - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Disk caching impacts performance.
> There is a physical constraint here. If you want to verify that your > data is safely written to non-volatile storage you have to live with the > latency. If that is unimportant to you you can relax the ACID > requirements and get faster writes, but when you do that there is no > crying over lost data after a crash. Hmmph. In my experience, there will be *lots* of crying over lost data after a crash, even if the risks are known and ostensibly accepted beforehand. :) Brad - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Disk caching impacts performance.
Ram Drive involve a context switch(from user to Kernel) and hence there is loss of performance!!! Check this factor also. regards ragha ** This email and its attachments contain confidential information from HUAWEI, which is intended only for the person or entity whose address is listed above. Any use of the information contained herein in any way (including, but not limited to, total or partial disclosure, reproduction, or dissemination) by persons other than the intended recipient(s) is prohibited. If you receive this e-mail in error, please notify the sender by phone or email immediately and delete it! * - Original Message - From: John Stanton <[EMAIL PROTECTED]> Date: Friday, November 9, 2007 7:02 pm Subject: Re: [sqlite] Disk caching impacts performance. > Michael Scharf wrote: > > [EMAIL PROTECTED] wrote: > >> "Trevor Talbot" <[EMAIL PROTECTED]> wrote: > >>> Beyond that, I'm not aware of anything that would help. > >>> > >> > >> All good advice. But you left off the obvious: Get a > >> faster disk drive. ;-) > > > > ...which does not really help unless you buy a very expensive > > flash disk drive. How much faster is a *really* fast spinning disk? > > > > Trevo, have you tried to put your database on a (fast!) USB stick. > > It should be much faster in 'seeking' but is slower in the > > data transfer. This would give some indication if the access > > is limited by seek or the disk reading speed. > > > > > > Michael > > > A USB flash drive is not particularly fast due to the limited write > speed of flash memory and buss speed. A fast disk spins at 15,000 > rpm, > double the speed of the higher end 7,500 rpm disks and almost 3 > times > the speed of the regular 5,400 rpm devices. > > If you want to simulate a disk with no latency set up a RAM drive. > > There is a physical constraint here. If you want to verify that > your > data is safely written to non-volatile storage you have to live > with the > latency. If that is unimportant to you you can relax the ACID > requirements and get faster writes, but when you do that there is > no > crying over lost data after a crash. > > > - > To unsubscribe, send email to [EMAIL PROTECTED] > > - > > - To unsubscribe, send email to [EMAIL PROTECTED] -
[sqlite] Problems after upgrade
Hi, I was using Sqlite library 3.3.8 before upgrading it to version 3.4.2 on Linux. Right after upgrade I noticed that the CPU load of my application, which is shared linked to the library, increased considerably. Profiling revealed that version 3.4.2 seems to have some kind of problem with the DB if created with version 3.3.8 of the library. These are the first lines of the profiled data on an (despite sqlite) basically idle system: samples %app name symbol name 574 7.1616 libsqlite3.so.0.8.6 sqlite3VdbeExec 485 6.0512 libsqlite3.so.0.8.6 __i686.get_pc_thunk.bx 315 3.9301 libsqlite3.so.0.8.6 sqlite3VdbeSerialType 265 3.3063 libsqlite3.so.0.8.6 sqlite3VdbeSerialPut 263 3.2813 libsqlite3.so.0.8.6 sqlite3PutVarint 228 2.8447 audld.plx(no symbols) 227 2.8322 libc-2.4.so _int_malloc 212 2.6450 libfreetype.so.6.3.8 (no symbols) 200 2.4953 vmlinux-2.6.16.43-54-default default_idle 182 2.2707 libcsscan.so convert_line 160 1.9963 libsqlite3.so.0.8.6 sqlite3PagerAcquire 149 1.8590 confd.plx(no symbols) ... Now, despite removing the old lib, is there some chance to keep the old DB and kind of "repair" it? Also, is there some kind of policy which tells me if the format of the DB changed? Thanks. /holger - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Disk caching impacts performance.
Renaud HUILLET wrote: Thanks for your reply, Indeed, the windows API is not the same at the Unix one (mmap), but I think I have a wrapper somewhere that can handle both. Anyone has been trying the mmap for SQLite ? Renaud Date: Thu, 8 Nov 2007 10:15:24 -0500 From: [EMAIL PROTECTED] To: sqlite-users@sqlite.org Subject: Re[2]: [sqlite] Disk caching impacts performance. Hello Renaud, Thursday, November 8, 2007, 9:11:41 AM, you wrote: RH> Is the mmap option part of sqlite or do we need to change the sqlite code ? RH> If so, has somebody here already tried it ? Any source available ? RH> What 's the order of improvement ? RH> Renaud Date: Thu, 8 Nov 2007 15:54:35 +0500 From: [EMAIL PROTECTED] To: sqlite-users@sqlite.org Subject: Re: [sqlite] Disk caching impacts performance. Try this, use mmap (i assume u'r db is readonly)this is much faster and better than ifstream read. Also ifstream read can keep the data in cache as long as no other serious i/o occurs. U need to accept it as we work with Virtual Mem or write your own FileSystem which is mem based and short circuits os calls. Sqlite 3.5x has good support for such ext. regrads ragha ** This email and its attachments contain confidential information from HUAWEI, which is intended only for the person or entity whose address is listed above. Any use of the information contained herein in any way (including, but not limited to, total or partial disclosure, reproduction, or dissemination) by persons other than the intended recipient(s) is prohibited. If you receive this e-mail in error, please notify the sender by phone or email immediately and delete it! * - Original Message - From: Julien Renggli Date: Thursday, November 8, 2007 4:15 pm Subject: [sqlite] Disk caching impacts performance. Hello, I'm currently working on a project which is very likely to use SQLite as data storage. Since performance is important for us, I already found on the SQLite website many ways to optimise the code (always working in transactions where possible, using a page size of 4096 since it's running on the Windows platform, using integers primary keys, ...). ButI have one problem that I "solved" in an unorthodox way; it works, but maybe you have a better solution than mine? I've been doing some test with a ~100 MB database, in which I have three tables: one for structured data (Objects, 2000 entries), one for the blobs we have to store (ObjectBlobs ID INTEGER PRIMARY KEY, Data BLOB) (8000 entries), and one which binds the structured data and the blobs (8000 entries). As you can imagine, each Object has 4 blobs linked to it; the blobs can be quite large (let's say up to 1 MB). My (C++) application just has to read the table "Objects", and one of the Blobs for each Object. Now the first time I run my application, it takes quite a long time (30s) to load the blobs. But if I re-run the app, it only takes 1s to load them. It's clearly a disk caching issue: if I copy huge files to the disk between two runs, it takes again 30s to load the blobs (i.e. the DB is no more in the disk cache). Profiling the application indicates sqlite::winRead() is the bottleneck. I then had the following idea: SQLite is probably reading the file randomly, depending on where the data lies. If I can force the DB to be cached, everything should be fine. So before connecting the database, I first read it sequentially (using a C++ ifstream) until the end of file. It perfectly solves the database problem, even though I still notice a difference (3s to read the file on 1st run, 0.2s later). But 3s is OK where 30s was worrying me. I hope I explain the situation clear enough, and ask you now: is it theonly way to do it? I find the trick a bit nasty and don't like it; maybe I missed something? Before you ask: I tried to VACUUM the DB, it only reduced the timings to 16s, which was still bad for our requirements. Tests with a larger DB (it can get much bigger than my example) and on different machines tend to confirm my theory. Thanks in advance (and a big thank for SQLite which is really nice and easy to use !), Julien Renggli P.S.: Some technical informations: - sqlite v.3.3.16 (we will upgrade to the latest version later), C API - Windows XP SP2 - Timings on Pentium 4 3.4GHz, 2GB RAM - To unsubscribe, send email to [EMAIL PROTECTED] - RH> _ RH> Explore the seven wonders of the world RH> http://search.msn.com/results.aspx?q=7+wonders+world=en-US=QBRE RH> - RH> To unsubscribe,
Re: [sqlite] Disk caching impacts performance.
Michael Scharf wrote: [EMAIL PROTECTED] wrote: "Trevor Talbot" <[EMAIL PROTECTED]> wrote: Beyond that, I'm not aware of anything that would help. All good advice. But you left off the obvious: Get a faster disk drive. ;-) ...which does not really help unless you buy a very expensive flash disk drive. How much faster is a *really* fast spinning disk? Trevo, have you tried to put your database on a (fast!) USB stick. It should be much faster in 'seeking' but is slower in the data transfer. This would give some indication if the access is limited by seek or the disk reading speed. Michael A USB flash drive is not particularly fast due to the limited write speed of flash memory and buss speed. A fast disk spins at 15,000 rpm, double the speed of the higher end 7,500 rpm disks and almost 3 times the speed of the regular 5,400 rpm devices. If you want to simulate a disk with no latency set up a RAM drive. There is a physical constraint here. If you want to verify that your data is safely written to non-volatile storage you have to live with the latency. If that is unimportant to you you can relax the ACID requirements and get faster writes, but when you do that there is no crying over lost data after a crash. - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] Suggests for improving the SQLite website
In general I'm agree with Kishor comments, specially those related to the ability to include comments to the documentation pages and the need of *more* examples. I recognize that the present design has its charm (those of a Web site made by an engineer) but I recognize that has found some difficulties to find some information, specially when beginning with the matter. Keeping in mind the type of public of this web, don't worry about certain language frank and direct . All us have the same problem (bugs) every day. As an additional suggestion for those of us who love SQLite but can't afford the cost of a maintenance subscription (or don't need it); put a "donate" button and add a list of contributors. A.J.Millan - To unsubscribe, send email to [EMAIL PROTECTED] -
Re: [sqlite] SQLite and Large Databases
Teg <[EMAIL PROTECTED]> wrote:
>
> dhc> Recompile version 3.5.2 with -DSQLITE_MEMORY_SIZE=1000
> dhc> or however much memory you want SQLite to use. This will
> dhc> create a static array of char[] of size 1000 (or whatever
> dhc> other size you give it) and use that instead of malloc()
> dhc> to obtain all the memory it needs. With this approach it
> dhc> is impossible for SQLite to use more than the specified
> dhc> amount of memory since it never calls malloc().
>
>
> Win32 here. I set this option and get insta-crashes in both debug and
> release builds. Using the latest 3.5.2 from the website.
>
> -DSQLITE_MEMORY_SIZE=1000
>
> It crashes right here when it was trying to push pager pages out.
>
Works OK when I do it.
> int sqlite3OsWrite(sqlite3_file *id, const void *pBuf, int amt, i64 offset){
> DO_OS_MALLOC_TEST;
> return id->pMethods->xWrite(id, pBuf, amt, offset);
> }
>
>
> If you need more information let me know. It's not causing my any real
> problems, I just disable the setting and it's back to normal.
>
>
> --
> Best regards,
> Tegmailto:[EMAIL PROTECTED]
>
>
> .
-
To unsubscribe, send email to [EMAIL PROTECTED]
-
Re: Re[2]: [sqlite] Disk caching impacts performance.
You can significant increase access speed by "ANALYZE" in some cases. -- Biomechanica Artificial Sabotage Humanoid
