Yes, I do this a lot ... dealing with diacriticals in books that have to go out to the web. Complete transformations of one font into another font (every single character gets replaced), fixing broken links in 100's of web pages... I have found absolutely nothing that can do this as efficiently as Revolution. Sometimes transforming characters across thousands of files. Operations that may take only 22 seconds across 3,000 web pages (on the LAN) in Revolution, would take hours in GoLive and probably GoLive would simply choke in the end. Because it is doing so much tracking and updating of it's internal data base of the webSite. I've been using my Rev App for 4 years and *never* had a single failure, bad write or corrupt file, not once.

First rule: Don't read data into a field.

Basic implementation is to get a list of paths to files that need work in one field called "fileList"

then run scripts against those files like this, where the fld ""sNr" is a series of lines with

[stringToReplace] tab [stringToChangeTo] in each line.

I keep a library of these "arrays" in a substack and pass them to fld "sNr"

writes to a new file, same name with galley # added.

on mouseUp
set the cursor to watch
put ticks() into tStart
put empty into fld "resultList"
answer "This operation will output a new file with " & quote & "2" & quote & cr &\
"appended to the file name. Continue?" with "No" or "Yes"
if it is "No" then exit MouseUp
set the caseSensitive to (the hilite of btn "Case sensitive")
set the itemdel to tab
put fld "sNr" into tChanges
put fld "fileList" into tFileList


  repeat for each line y in tFileList
    set the itemdel to tab

    put url ("file:"&y) into tFileData

    repeat for each line z in tChanges
      replace (item 1 of z)  with (item 2 of z) in tFileData
    end repeat

    if the hilite of btn "Write New File" then
    set the itemdel to "."
    put item 1 of y &"-g2" & ".txt" into x
    put tFileData into  url ("file:"&x)
    else
    put tFiledata into url ("file:"&y)
    end if

end repeat
  calcTime(tStart)
end mouseUp

It's  actually incredibly simple.

If you want the stack, let me know off line..it's a bit of a mess but works great.

Sivakatirswami







On Mar 26, 2005, at 1:27 PM, Andre Garzia wrote:

Hi Folks,

I am having some trouble escaping text with international characters for HTML pages. I thought a simple approach, I built a field with all the escapes like:

á á
é é
Á Á
...

and stuff like that, then I used a repeat loop to loop thru the data replacing the ocurrence of the first char with the corresponding HTML entity, guess what? It does not work, the replace command is not replacing them, it's a big field, it works only for some lines, others are completelly ignored, I used the debugger to iterate thru the code, and guess what, it's all appearing correct just the engine is telling the the "é" inside the field is not the same "é" that it needs to escape... damn. Any clue out there? does anyone here got a HTML escaping routines for international chars?

Andre
--
Andre Alves Garzia  2004  BRAZIL
http://studio.soapdog.org

_______________________________________________
use-revolution mailing list
[email protected]
http://lists.runrev.com/mailman/listinfo/use-revolution


_______________________________________________ use-revolution mailing list [email protected] http://lists.runrev.com/mailman/listinfo/use-revolution

Reply via email to