On Jun 7, 2018, at 2:32 PM, Stephen Chrzanowski <pontia...@gmail.com> wrote:
> 
> On Thu, Jun 7, 2018 at 4:07 PM, Warren Young <war...@etr-usa.com> wrote:
> 
>> Do you really need something in the daily scrape that you wouldn’t get
>> from a Fossil clone?
> 
> To that, I can honestly say, I don't know.  The thing I like about the
> daily scrapes is that I can grab the zip files provided in whatever state
> they were uploaded as, and I archive them so 'just in case' I can go back
> to a particular DLL and see whats going on.

If you’re talking about the contents of

    https://sqlite.org/download.html

then I believe that is not stored in Fossil, as is good version control 
hygiene, those all being artifacts derived from source files that *are* checked 
into the repository.  If the DLLs and such were checked into the Fossil 
repository, you’d expect to find them here:

    https://www.sqlite.org/cgi/src/uvlist

But that feature of Fossil is apparently not being used by the SQLite project 
at the moment.

Since those are all generated from files that are in Fossil, however, that 
means you can also build them in the same way:

    fossil clone https://sqlite.org/cgi/src ~/museum/sqlite.fossil
    mkdir ~/sqlite-src
    cd ~/sqlite-src
    fossil open ~/museum/sqlite.fossil version-3.22.0
    ./configure
    make sqlite3.c
    x86_64-w64-mingw32-gcc -shared sqlite3.c -o sqlite3.dll

Notes:

0. You will need a recent version of Fossil to clone SQLite.  It appears Fossil 
was never packaged for Lenny, nor would it work if it did due to a 
cryptographic hashing change made on the repository last year.  If the official 
binaries won’t run on Lenny, it’s easy enough to build Fossil from source:

    https://fossil-scm.org/index.html/uv/download.html

Or:

    wget --no-check-certificate 
https://fossil-scm.org/index.html/uv/fossil-src-2.6.tar.gz
    tar xvf… && ./configure && make && sudo make install

1. ~/museum is where I choose to keep my fossils.  Fossil doesn’t care where 
they are, nor must they be grouped in any particular way.  Use the organization 
scheme that makes sense to you if you don’t like mine.

2. Fossil also doesn’t care that the repository clones end in *.fossil, with 
one exception you’re not likely to run into unless you get much deeper into 
Fossil.  Still, it’s a good convention to follow.

3. After doing the initial repository clone, you don’t need to re-clone it each 
time you want a new version.  You just need to say “fossil update” within a 
checkout directory whenever you want to fetch upstream changes.  Some 
operations will automatically do an update — technically called an autosync — 
so you might not have to do it manually.

4. You only need to “open” the repository into a different checkout directory 
each time if that’s helpful to you, as it can be whenever it’s useful to have 
multiple branches open at once.  If you want just one checkout directory, after 
the initial “open” you can switch to saying “fossil update version-xxxx” in 
place of the “open” command.

5. There are many other things you can give in place of the version tag: 

    https://www.sqlite.org/cgi/src/brlist
    https://www.sqlite.org/cgi/src/taglist
    https://www.fossil-scm.org/xfer/doc/trunk/www/checkin_names.wiki

You might, for example, find it useful to fetch a version as of a particular 
date:

    fossil update 2017-07-04

Or, get the latest release, without having to know in advance what the version 
number is:

    fossil update release

6. The final three steps in the build sequence above assume you’re doing this 
on the Debian box with the MinGW 64-bit cross-compiler:

    http://archive.debian.org/debian/pool/main/m/mingw-w64/

If you need a 32-bit binary, I believe you just need to switch the compiler 
command to i686-w64-mingw32-gcc.  (That’s a guess.  I’ve never had to use the 
MinGW cross-compiler on a Linux box.  I rarely enough use it on Windows.)

If you do this on a Windows box instead, the build instructions differ:

    https://sqlite.org/howtocompile.html#building_a_windows_dll
_______________________________________________
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users

Reply via email to