It was on headless server using *grep***... I have parsed all
resourcestrings from lazarus and then was looking for their appearance
in the code... I just wrote short script in bash, because I had to write
it through ssh from mobile (free time, no laptop). I have tested search
in files function on the server and it took 3 seconds first time to
search through whole source tree... I'm sorry if I made any confusion or
gossip around lazarus Search in Files function :-D
Václav Valíček
[email protected]
Dne 16.7.2014 20:18, Mattias Gaertner napsal(a):
On Wed, 16 Jul 2014 18:36:23 +0200
Václav Valíček <[email protected]> wrote:
Same overkill as you describe above, if you realise that most projects
are dependent on LCL (And it on LCLBase, next LazUtils, next FCL...) and
it is huge bunch of files....
Once searched, the OS has the files in cache and search takes
only a few seconds on recent machines.
While I was doing experiments with resourcestrings, I had clean git
checkout of repo (without binaries) and script that greped
resourcestrings (or any other word) in source tree... On quadcore server
with 4GB of RAM, WD Black (server) HDD and only running ssh and nginx
(load was ~0.15) the search took about 25-30 seconds... On my laptop, it
was two minutes...
What have you searched?
Using "Find in files" to search for a plain string case insensitive in
the whole Lazarus sources (no fpc sources) *.pas;*.pp;*.inc;*.lfm under
Linux takes about 3 secs here.
Of course the first time is much longer, when the OS has to load the
files from disk.
Mattias
--
_______________________________________________
Lazarus mailing list
[email protected]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
--
_______________________________________________
Lazarus mailing list
[email protected]
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus
