This tool tries to fill in the wet French document after the wet English document has been manually completed. It works on the assumptions that a) the original (non-wet) documents have similar document structures and b) the wet html documents should have identical structures. With that in mind, it replaces the English values in the wet html document with their equivalent French values.
If the output is especially strange (e.g. the entire document being unmatched), try clicking through the buttons more slowly (some of this is asynchronous).
Steps (examples and so on can be found in the readme):
Part 1 - get contents from dirty documents
In Word, stop tracking changes and delete comments.
Mark any superscripts or subscripts in the same manner as for the general Dreamweaver formatting tool (I use the same code for some initial cleanup).
Paste these Word documents as HTML files into the Dreamweaver design view in separate files.
Upload these two HTML files.
Download the values that are generated (these are the html documents with the tags removed).
Part 1.5 - making sure the contents align -
Make sure there are exactly the same number of English and French values (number of values is printed in the console - ctrl+shift+i) since the next part matches by index.
If they don't, manually align the rows so that each line of English content is on the same row as the corresponding French content.
Some tips if you are using Beyond Compare to align values:
Go to Session -> Session Settings -> Alignment and set the values to Unaligned so that the rows are aligned by index.
Make sure to save and reload after every change you make so that you can visually keep track of how the rows are aligned (unless you can follow it on your own), since each of your edits will dealign the rows.
The alignment process is as follows:
If the English document has extra rows of content from having tags that the French document lacks, add blank rows to the French content so that its contents align with the English content before and after the extra rows.
If the English document has extra rows of content from its content coming before or after a tag differently, add blank rows to the French content until their rows align.
If the French document has extra rows of content from having tags that the English document lacks, add lines in the English content beginning with < and followed by the tag that separates this row of French content from the next row, so that the remaining content aligns.
If the French document has only one extra row of content instead of two because its extra tag is next to another tag, end the tag on the equivalent line in the English content with >>.
If the French document has an extra row of content that should be appended to a brand new line, surround the tag on the equivalent line in the English content with <? and ?>.
If the French document has extra rows of content from its content coming before or after a tag differently:
Add a row with the placeholder tag <!r in the English contents to indicate that the equivalent French row should be pushed one tag behind, <!r2 to push it two tags behind, or <!r3 to push it three tags behind.
Add a row with the placeholder tag <!l in the English contents to indicate that the equivalent French row should be pushed one tag in front, <!l2 to push it two tags in front, or <!l3 to push it three tags in front.
If the documents have different tags at the same index but are otherwise aligned, you can treat it as the English document having extra rows followed by the French document having extra rows.
Part 2 - replace english contents in cleaned english document with french contents
Upload the values after you've made sure they align.
Upload the English document that was formatted by WET standards.
Download the list of unmatched English values, and then the French document that should now be formatted by WET standards.
Go through the list of unmatched English values and, if applicable, add them into the French document manually.
If the option to check for math-ignoring and math-reducing matches is selected, inline math will not prevent a match, but the math will be mispositioned to the start of the parent tag. Manually go through <math> tags in the cleaned French document and reposition them correctly. If you use Dreamweaver, I recommend doing this before applying source formatting so that the math doesn't get moved to its own line.
Make sure to double check with beyond compare as usual. You may want to manually search through and remove problematic cases of the following:
Empty tags generated by blank french content lines. Regex search string: <[^/>]*> *</[^>]*>
Internal tags that don't have spacing around them (usually from inserted tags into the English contents). Regex search string: [a-zA-Z0-9]<[^>]*>[a-zA-Z0-9]
More in-depth information can be found in the readme.