Re: [fossil-users] is there a push/sync/commit hook for unit testing ?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 On 21.01.2011 21:36, Nolan Darilek wrote: > On 01/21/2011 07:44 AM, Richard Hipp wrote: >> will be appreciated. Though, I suppose I could make these things a >> posix-only feature. > > > That would be my vote, if in fact you are interested in taking votes. :) My vote you have too on that point ... so we are now 2 ... :-) > No sense in *everyone* missing out because Windows isn't Posix, and if > you or someone else finds out how to run background tasks under Windows > one day then they can have it too. > > Really looking forward to hooking some Fossil projects into Hudson. > Hooks are the one thing I keep finding myself missing from Git over the > past few months. Otherwise I haven't looked back. I thought about trying > something that watched a directory of fossils for changes and pulled the > RSS feed for what's new, but hooks would be so much better than that. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iF4EAREIAAYFAk0580AACgkQPKy7P5mnYdkFuwD9FMlXhqSllNQi/1AszWyuoL6f +AAIhi/XPhYzbt0ZNa0A/1tLTJ0FcZJjne5URcoT3/gIzBdm6Z2nW3geevU1OLU3 =yAKH -END PGP SIGNATURE- ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] is there a push/sync/commit hook for unit testing ?
On Jan 21, 2011, at 1:45 PM, Richard Hipp wrote:
> But I did not accept that branch onto the trunk. I want to do the "hooks"
> via a different mechanism. But in order to implement this different
> mechanism, I need example C code for launching a background process in
> windows that is not associated with an console. Doing that is easy in Unix,
> but I'm not sure how to do it in windows.
>
> This background process would be responsible for waiting until the push/pull
> finished and then running whatever hooks are required.
I haven't programmed on Windows for ages, and maybe it's not what you want, but
here's some code I tried.
Basically, there are two interesting creation flags for CreateProcess:
CREATE_NEW_CONSOLE and DETACHED_PROCESS. The former opens a new DOS console for
a child process, the latter doesn't. DETACHED_PROCESS seems like what you want,
but I cannot make it write into the same console from where the parent process
launched.
The following is an example with modified win32_create_child_process from
popen.c. When you launch it, parent process prints "Hello from parent", creates
a child process and exits. Child continues running even if you close "DOS"
window. It will write "Hello, I'm a child process, writing #[number of write]"
to file C:\process-test.txt every second until you close it with Task manager.
#include
#include
#include
static int win32_create_child_process(
char *zCmd, /* The command that the child process will run */
HANDLE hIn, /* Standard input */
HANDLE hOut, /* Standard output */
HANDLE hErr, /* Standard error */
DWORD *pChildPid /* OUT: Child process handle */
){
STARTUPINFO si;
PROCESS_INFORMATION pi;
BOOL rc;
memset(&si, 0, sizeof(si));
si.cb = sizeof(si);
si.dwFlags = STARTF_USESTDHANDLES;
SetHandleInformation(hIn, HANDLE_FLAG_INHERIT, TRUE);
si.hStdInput = hIn;
SetHandleInformation(hOut, HANDLE_FLAG_INHERIT, TRUE);
si.hStdOutput = hOut;
SetHandleInformation(hErr, HANDLE_FLAG_INHERIT, TRUE);
si.hStdError = hErr;
rc = CreateProcess(
NULL, /* Application Name */
zCmd, /* Command-line */
NULL, /* Process attributes */
NULL, /* Thread attributes */
TRUE, /* Inherit Handles */
DETACHED_PROCESS, /* Create flags */
NULL, /* Environment */
NULL, /* Current directory */
&si, /* Startup Info */
&pi/* Process Info */
);
if( rc ){
CloseHandle( pi.hProcess );
CloseHandle( pi.hThread );
*pChildPid = pi.dwProcessId;
}else{
//win32_fatal_error("cannot create child process");
printf("cannot create child process: %d", rc);
exit(0);
}
return rc!=0;
}
int main(int argc, const char *argv) {
if( argc > 1 ){
int i = 0;
for(;;){
FILE *f = fopen("C:\\process-test.txt", "a");
fprintf(f, "Hello, I'm a child process, writing #%d\n",
i++);
fclose(f);
Sleep(1000);
}
}else{
DWORD pChildPid;
printf("Hello from parent\n");
win32_create_child_process("a.exe child",
GetStdHandle(STD_INPUT_HANDLE),
GetStdHandle(STD_OUTPUT_HANDLE), GetStdHandle(STD_ERROR_HANDLE),
&pChildPid);
}
}
--
Dmitry Chestnykh
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] is there a push/sync/commit hook for unit testing ?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 01/21/2011 07:44 AM, Richard Hipp wrote: > will be appreciated. Though, I suppose I could make these things a > posix-only feature. > That would be my vote, if in fact you are interested in taking votes. :) No sense in *everyone* missing out because Windows isn't Posix, and if you or someone else finds out how to run background tasks under Windows one day then they can have it too. Really looking forward to hooking some Fossil projects into Hudson. Hooks are the one thing I keep finding myself missing from Git over the past few months. Otherwise I haven't looked back. I thought about trying something that watched a directory of fossils for changes and pulled the RSS feed for what's new, but hooks would be so much better than that. -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk057lAACgkQIaMjFWMehWKIQwCeMm9OWuQSet537luMdIstFqj8 glIAn2KHinO4fYn18wFedLIO2muZXJDQ =Y6z0 -END PGP SIGNATURE- ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] is there a push/sync/commit hook for unit testing ?
On Fri, Jan 21, 2011 at 10:23 AM, Ron Wilson wrote: > Why not use threads? Perhaps you said this because you are aware of my opinion of threads and you just wanted to push my buttons? Ha! I'm on to your ruse and it won't work! > Multi-threaded processes are supported on both > Windows and POSIX. > > Also, it is poosible to catch the Control-C signal in your process so > you can wait until in-progress operations are complete before > terminating the process. Your Control-C handler should output a > message (to STDERR) like "Shutdown waiting for in-progress operations > to complete." > In the case of CGI, the background process may need to wait until one or more additional messages are received by subsequent client messages. But the client won't send any more messages until the current CGI returns. And the current CGI can't return until the background thread completes. So you are deadlocked. > > On Fri, Jan 21, 2011 at 8:44 AM, Richard Hipp wrote: > > Actually, it would even be OK to kill the > > process when the DOS box is closed, as long as the process continues > running > > after the user presses Ctrl-C to shut down the "fossil server" command. > > I suppose the most important thing is that I need the background process > to > > continue running after the parent process (which will usually be the CGI > on > > the server side of a push/pull) returns its output to the webserver > and/or > > user and terminates. > > A push/pull typically involves multiple round-trips between client and > > server. The background process is needed so that we can wait until the > > entire push/pull operation completes before invoking the hooks. The > > background process is also needed so that a long-running hook does not > delay > > the server from responding to the client. > ___ > fossil-users mailing list > fossil-users@lists.fossil-scm.org > http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users > -- D. Richard Hipp d...@sqlite.org ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Diff highlighting
On 01/21/2011 07:57 AM, Rene wrote:
> I think that we better could do
>
> This has the benefit that if someone else want to do something on
> load he could add it in the init function.
It is not necessary to codify one inline init function--JavaScript
programs can share event handlers without having any knowledge of each
other. (Admittedly it's a little verbose, but it works.) Below is the
basic JavaScript pattern for adding something to an event handler
without clobbering any existing event handlers and without polluting the
global namespace. (It doesn't guarantee that no later script will
clobber this one, but nothing can guarantee that.)
(function(){
var myInit = function(){
// do stuff
};
if(window.onload) {
var oldOnload = window.onload;
window.onload = function(){
oldOnload.apply(this,arguments);
myInit.apply(this,arguments);
}
} else {
window.onload = myInit;
}
})()
--
Joshua Paine
LetterBlock: Web applications built with joy
http://letterblock.com/
301-576-1920
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] is there a push/sync/commit hook for unit testing ?
Why not use threads? Multi-threaded processes are supported on both Windows and POSIX. Also, it is poosible to catch the Control-C signal in your process so you can wait until in-progress operations are complete before terminating the process. Your Control-C handler should output a message (to STDERR) like "Shutdown waiting for in-progress operations to complete." On Fri, Jan 21, 2011 at 8:44 AM, Richard Hipp wrote: > Actually, it would even be OK to kill the > process when the DOS box is closed, as long as the process continues running > after the user presses Ctrl-C to shut down the "fossil server" command. > I suppose the most important thing is that I need the background process to > continue running after the parent process (which will usually be the CGI on > the server side of a push/pull) returns its output to the webserver and/or > user and terminates. > A push/pull typically involves multiple round-trips between client and > server. The background process is needed so that we can wait until the > entire push/pull operation completes before invoking the hooks. The > background process is also needed so that a long-running hook does not delay > the server from responding to the client. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] is there a push/sync/commit hook for unit testing ?
On Fri, Jan 21, 2011 at 07:45:39AM -0500, Richard Hipp wrote: > But I did not accept that branch onto the trunk. I want to do the "hooks" > via a different mechanism. But in order to implement this different > mechanism, I need example C code for launching a background process in > windows that is not associated with an console. Doing that is easy in Unix, > but I'm not sure how to do it in windows. We write some wxWidgets programs, and we can call other executables without trouble through wxProcess/wxExecute, without any console appearing. Maybe you can check the wxWidgets source code of wxProcess/wxExecute for the details http://docs.wxwidgets.org/stable/wx_wxprocess.html Source code related for Windows: http://trac.wxwidgets.org/browser/wxWidgets/branches/WX_2_8_BRANCH/src/msw/utilsexc.cpp Regards, Lluís. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Fossil status icons in TextMate
Hello, I've added Fossil support to ProjectPlus plugin for TextMate (it shows added/deleted/etc icons for files in the project drawer). I've send pull request to the author, so I hope it will be included in the next release of the plugin. For now, the source code is here: https://github.com/dchest/projectplus Unfortunately, I couldn't compile the related SVN libraries, so my build comes with Fossil support, but without SVN support (if anyone can do it, please let me know). The build is here: http://dl.dropbox.com/u/126472/Fossil/ProjectPlus.tmplugin.zip (I haven't tested it on PPC, but it should work on both i386 and PPC). Don't forget to turn on Fossil in TextMate > Preferences > ProjectPlus > SCM. By default it looks for fossil binary in /opt/local/bin, /usr/local/bin, or /usr/bin. PS May I also remind Mac users about QLFossil: http://codingrobots.org/p/qlfossil/ -- Dmitry Chestnykh ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] is there a push/sync/commit hook for unit testing ?
On Fri, Jan 21, 2011 at 8:24 AM, Francisc Simon wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA256 > > The only way, in Windows, that you can have a process started by a user > continue running after logoff (i.e. what "nohup" does) is to start it > either through a "scheduled task" or as a Windows service. When the user > logs off all processes in their logon session will be killed. > > I don't need the process to continue running after the user logs off - I only need it to keep running after the user closes the DOS box in which she had typed "fossil server". Actually, it would even be OK to kill the process when the DOS box is closed, as long as the process continues running after the user presses Ctrl-C to shut down the "fossil server" command. I suppose the most important thing is that I need the background process to continue running after the parent process (which will usually be the CGI on the server side of a push/pull) returns its output to the webserver and/or user and terminates. A push/pull typically involves multiple round-trips between client and server. The background process is needed so that we can wait until the entire push/pull operation completes before invoking the hooks. The background process is also needed so that a long-running hook does not delay the server from responding to the client. Among the things I want to do with this same "hook" mechanism is a relay-push. When server A receives a push, it will automatically relay the new content to server B. For example, when you push to http://www.fossil-scm.org/ it will automatically relay that content to http://www2.fossil-scm.org/ and http://www3.fossil-scm.org/. Currently, that relay is accomplished using cron, but as you have already pointed out, cron is a complication we would like to avoid. Another feature that might be achieved using the background process is email notification of changes. This would be a nice-to-have, though is less important since we already have an RSS feed built in. But notice that we need a background process to accomplish this because we don't want the server response to be delayed while it sends the emails. This is all relatively easy to do on posix. Hints for doing it on windows will be appreciated. Though, I suppose I could make these things a posix-only feature. -- D. Richard Hipp d...@sqlite.org ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] is there a push/sync/commit hook for unit testing ?
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
On 21.01.2011 13:45, Richard Hipp wrote:
> On Thu, Jan 20, 2011 at 11:52 PM, Francisc Simon
> wrote:
>
>
> Hi @all,
>
> i probably ask a question that was already asked before but i can not
> figure out from the tickets/mailing-list cache if this was already
> solved or not.
>
> Is there a way to hook into push/sync/commit commands ?
>
>
>> There is a branch in the source tree that attempts to do this. See, for
>> example
>
>> http://www.fossil-scm.org/fossil/timeline?r=StvPrivateHook2+
>
>> And especially
>
>> http://www.fossil-scm.org/fossil/ci/ba5e03444b
>
>> But I did not accept that branch onto the trunk. I want to do the "hooks"
>> via a different mechanism. But in order to implement this different
>> mechanism, I need example C code for launching a background process in
>> windows that is not associated with an console. Doing that is easy in Unix,
>> but I'm not sure how to do it in windows.
>
>> This background process would be responsible for waiting until the push/pull
>> finished and then running whatever hooks are required.
>
>
Ok, i found on the web the following solution:
http://serverfault.com/questions/39451/there-is-any-way-to-run-processes-in-the-background-in-windows-nohup-equivalent
The only way, in Windows, that you can have a process started by a user
continue running after logoff (i.e. what "nohup" does) is to start it
either through a "scheduled task" or as a Windows service. When the user
logs off all processes in their logon session will be killed.
If you're game to try the "Scheduled Tasks" method, you'll want to know
how to create them programmatically. The *Win32_ScheduledJob* WMI class
can do it. Documentation is provided in detail here:
http://www.microsoft.com/technet/scriptcenter/guide/sas_man_rsxs.mspx?mfr=true
Basically, you're looking at doing (shamelessly stolen from Microsoft):
Set objService = GetObject("winmgmts:\\.")
Set objNewJob = objService.Get("Win32_ScheduledJob")
errJobCreated = objNewJob.Create("Program-to-execute.exe",
"ugly-formatted-time-string-per-Microsoft-docs",True ,1 OR 4 OR 16, ,
,JobID)
If Err.Number = 0 Then
Wscript.Echo "New Job ID: " & JobID
Else
Wscript.Echo "An error occurred: " & errJobCreated
End If
To grant "joe user" the ability to create scheduled tasks, you'll have
to modify the permission on the %SystemRoot%\Tasks folder. See here for
some info on that front:
http://technet.microsoft.com/en-us/library/cc785125(WS.10).aspx
i do not know if this will help you, i'm not a windows developer and i
guess i will never be one :-)
On the other hand is there a reason why you do not want to add first the
unix hooks and after that if even possible the windows one ?
The most of us even microsoft i guess use *NIX servers for production.
>
>
> reason: i need a hook that i can run the unit tests for my projects i
> use the check unit test framework and it has a nice feature it produces
> XML based logfiles of the tests running. This files can be displayed so
> i do not need to install a continous integration server :-)
> It would be nice to have a hook because i'm not a very big fan of cron
> jobs... :-)
>
> Thx for help!
>
> Regards
> Frank
>
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
>>
> ___
> fossil-users mailing list
> fossil-users@lists.fossil-scm.org
> http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Thx for your time!
Regards
Frank
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
iF4EAREIAAYFAk05iRsACgkQPKy7P5mnYdnhKgD/bz86rbcEykzi0qlMV8Z1RUKM
uh5AF2SH2KWe+HbpZosBAI5Dz8pGFzsDOjiRhJvxUNsmR/WI0bSCJk6nXLjdqVlt
=/zPJ
-END PGP SIGNATURE-
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Diff highlighting
On Fri, 21 Jan 2011 13:44:47 +0100, Dmitry Chestnykh wrote:
> On Jan 21, 2011, at 1:30 PM, Rene wrote:
>
>> You could define every thing in the head and do
>>
>> >
>> if { "fdiff" eq $current_page || "otherdiffed page" eq
>> $current_page
>> } {
>>html "
>> onload='DiffHighlighter.highlightElementsWithTagName(\'pre\')'"
>> }
>>
>>>
>>
>> This would have the benefit of running highlighter only on diff
>> pages
>> and after the page has loaded
>
> Awesome. The only downside is that onload will wait for images to
> load
> (http://www.javascriptkit.com/dhtmltutors/domready.shtml)
> but it doesn't matter that much.
>
> Does anyone has the list of all pages that have diffs?
>
> --
> Dmitry Chestnykh
>
> ___
> fossil-users mailing list
> fossil-users@lists.fossil-scm.org
>
> http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
I think that we better could do
function init()
{
if { "fdiff" eq $current_page || "otherdiffed page" eq
$current_page)
puts "DiffHighlighter.highlightElementsWithTagName('pre')"
}
This has the benefit that if someone else want to do something on load
he could
add it in the init function.
I did a egrep "WEB.*diff"on the files in the src directory and found
info.c:** WEBPAGE: vdiff
info.c:** WEBPAGE: fdiff
wiki.c:** WEBPAGE: wdiff
--
Rene
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] is there a push/sync/commit hook for unit testing ?
On Thu, Jan 20, 2011 at 11:52 PM, Francisc Simon wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA256 > > > Hi @all, > > i probably ask a question that was already asked before but i can not > figure out from the tickets/mailing-list cache if this was already > solved or not. > > Is there a way to hook into push/sync/commit commands ? > There is a branch in the source tree that attempts to do this. See, for example http://www.fossil-scm.org/fossil/timeline?r=StvPrivateHook2+ And especially http://www.fossil-scm.org/fossil/ci/ba5e03444b But I did not accept that branch onto the trunk. I want to do the "hooks" via a different mechanism. But in order to implement this different mechanism, I need example C code for launching a background process in windows that is not associated with an console. Doing that is easy in Unix, but I'm not sure how to do it in windows. This background process would be responsible for waiting until the push/pull finished and then running whatever hooks are required. > > reason: i need a hook that i can run the unit tests for my projects i > use the check unit test framework and it has a nice feature it produces > XML based logfiles of the tests running. This files can be displayed so > i do not need to install a continous integration server :-) > It would be nice to have a hook because i'm not a very big fan of cron > jobs... :-) > > Thx for help! > > Regards > Frank > > -BEGIN PGP SIGNATURE- > Version: GnuPG v1.4.10 (GNU/Linux) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ > > iF4EAREIAAYFAk05EPAACgkQPKy7P5mnYdlZBwD9HgczgCOeIOfkmOAtUdAw5KEY > +RxvpjBYPGJgN5/t/qsBAI88VZhhgB87ouhpxBu6SKrV4FudV0qFWVzMsbxs/us4 > =jwW6 > -END PGP SIGNATURE- > ___ > fossil-users mailing list > fossil-users@lists.fossil-scm.org > http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users > -- D. Richard Hipp d...@sqlite.org ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Diff highlighting
On Jan 21, 2011, at 1:30 PM, Rene wrote:
> You could define every thing in the head and do
>
>
> if { "fdiff" eq $current_page || "otherdiffed page" eq $current_page
> } {
>html "
> onload='DiffHighlighter.highlightElementsWithTagName(\'pre\')'"
> }
>
>>
>
> This would have the benefit of running highlighter only on diff pages
> and after the page has loaded
Awesome. The only downside is that onload will wait for images to load
(http://www.javascriptkit.com/dhtmltutors/domready.shtml)
but it doesn't matter that much.
Does anyone has the list of all pages that have diffs?
--
Dmitry Chestnykh
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Diff highlighting
On Thu, 20 Jan 2011 20:29:40 +0100, Dmitry Chestnykh wrote:
> Hello,
>
> I've made a tiny simple diff highlighting for Fossil (well,
> technically, for any HTML page) in JavaScript, and thought I'd share
> it with you.
>
> Just put the following somewhere into Footer (not header!) above
> :
>
>
>
>
> /* Simple diff highlighting */
> var DiffHighlighter = {
>
> isDiff : function(s){
> return (s.match(/^@@.*@@/m) && s.match(/^[+-]/m));
> },
>
> highlightElement : function(el){
> var s = el.innerHTML;
> if (!this.isDiff(s)){
> return;
> }
> el.className += 'diff';
> s = s.replace("<", "<");
> s = s.replace(/^\+.*$/mg, '$&');
> s = s.replace(/^\-.*$/mg, '$&');
> s = s.replace(/^@@.*$/mg, '$&');
> s = "" + s + "
"; // workaround for IE
> el.innerHTML = s;
> },
>
> highlightElementsWithTagName : function(tagName){
> var els = document.getElementsByTagName(tagName);
> for (var i=0; i < els.length; i++){
> this.highlightElement(els[i]);
> }
> }
> };
>
> DiffHighlighter.highlightElementsWithTagName('pre');
>
>
>
>
> And add this (or something to your taste) to your CSS:
>
>
>
> .diff-mark {
> display: inline;
> color: #000;
> }
>
> .diff-mark.position {
> display:-moz-inline-stack;
> display:inline-block;
> zoom:1;
> *display:block;
> width: 100%;
> font-style: italic;
> padding: 0.5em 0;
> margin: 0.5em 0;
> border-top: 1px dotted #A2B5CD;
> border-bottom: 1px dotted #A2B5CD;
> color: #A2B5CD;
> }
>
> .diff-mark.added {
> background-color: #CEFBC3;
> }
>
> .diff-mark.removed {
> background-color: #F5C2C1;
> }
>
>
>
> That's it: the script will automatically detect diffs and color them.
>
> You can see it in action here:
> http://codingrobots.org/p/fossil/ci/7ad9a4640a
>
> --
> Dmitry Chestnykh
>
> ___
> fossil-users mailing list
> fossil-users@lists.fossil-scm.org
>
> http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
You could define every thing in the head and do
if { "fdiff" eq $current_page || "otherdiffed page" eq $current_page
} {
html "
onload='DiffHighlighter.highlightElementsWithTagName(\'pre\')'"
}
>
This would have the benefit of running highlighter only on diff pages
and after the page has loaded
--
Rene
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] ignore list
On Tue, 25 May 2010 08:43:22 -0400 Richard Hipp wrote: > See the "ignore-glob" setting. Coming from other DVCS (darcs, bzr..) I am accustomed to have kind of 'default' ignore-glob list catching most often used file suffixes. What do you think about providing some 'default' list so that Fossil user should not start with adding e.g. *.*~ to the list? Sincerely, Gour -- Gour | Hlapicina, Croatia | GPG key: CDBF17CA signature.asc Description: PGP signature ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Eclipse IDE
Hi, I thought I heard somebody long ago working on an Eclipse plugin for Fossil. Anybody knows ? Regards, Paolo Lulli ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
