Closed GoogleCodeExporter closed 9 years ago
Update 1:
Tested alone, there is no problem with the length of one substitution, I've
tested the largest alone and worked.
Also there is no problem with the length of the entire file, because I decrease
it trimming some substitions and also doesn't work.
I guess, may be a special char could be the problem, I'll try to find any
regular expression within the text.
thanks
Original comment by glourenc...@gmail.com
on 3 Dec 2011 at 3:06
all set, there was an "&" inside the input text, spliting input in 2 lines
without the "&" was all set.
there would be nice to explain that '&' character may cause trouble.
ready to close
thanks
Ggustavo
Original comment by glourenc...@gmail.com
on 3 Dec 2011 at 5:04
finally, in the help, the link to "regular expresions" explanation in mozilla
help doesn't show the right explanation
Original comment by glourenc...@gmail.com
on 3 Dec 2011 at 5:07
Hi, nice to see you could resolve the problem for yourself ;)
The issue with the ampersand is not in the regular expressions of JavaScript,
but in the XML format itself. '&' marks the start of an entity, as wells as ';'
marks the end. So, to use represent '&' you have to put '&' in the XML file.
For other predefined entities you can refer to
https://en.wikipedia.org/wiki/List_of_XML_and_HTML_character_entity_references#P
redefined_entities_in_XML
Original comment by marc.r...@gmail.com
on 4 Dec 2011 at 1:29
Hey, it's sunday! many thanks for your clear explanation. FoxReplace it's
really incredible.
I'll tell my experience for your records. Now, working with an XML file which
has near 5000 reusable translatedc phrases and 950 kbytes. With this large file
"whole words" is faster than "text". Using "text" literally dies, with "whole
words" take about 5 to 10 seconds (using a quad core with 4gb ram and over
windows 7 64 bits with nightly firefox 11.0, this is a bit faster than firefox
8). Try to discover if something may things faster, only one cpu core is
allocated during F2.
regards
Gustavo
Original comment by glourenc...@gmail.com
on 4 Dec 2011 at 2:40
I didn't know that "whole words" is faster than "text" (at least in your case),
and now I don't know why. It could be that RegExp search is faster than plain
string search, as "whole words" uses a RegExp and "text" doesn't, but I'll have
to try.
Also, all the code is single-threaded, so it can only use one core, but I'll
investigate if it can be multi-threaded.
I take note of both ideas. Thanks!
Original comment by marc.r...@gmail.com
on 4 Dec 2011 at 7:01
Original issue reported on code.google.com by
glourenc...@gmail.com
on 3 Dec 2011 at 2:03