I've posted a beta version of the Language Reference Parser
online at my web site (www.software-in-motion.com).

Thanks Scott. I downloaded and ran it with only one obvious issue:
unlike the output file in the download, mine doesn't have any line
breaks or indentation.

Some suggestions:

- Add a ReadMe file that includes the "handles/does not handle" info.
- Add a version number, if only on the ReadMe file.
- List the version/platform of RB you usually use (both for compiling
and for extraction).
- Add an example of the output being used in some kind of (possibly
web) browser.
- In LRParser.LocateLRFiles, don't subtract one from ListCount; this
is not robust. Instead, special-case "Missing.html" during processing.
- Complete the separation of engine and UI by moving
LRParser.LocateLRFiles and LRParser.LocateLRTopicsFolder to the main
window. (This would make the LRParser module easier to re-use
elsewhere. For example, it could be dropped into UTool, which has its
own file/folder selection UI.) I would give LRParser a constructor
that took an RB application package FolderItem as a parameter, and I
would have it send an event for each file processed, so the main
window (or any other caller) can update the UI appropriately.
- Add a diff function so that XML output can be compared to a reference version.

What do you want to happen next? In particular, are you interested in
special-casing the source errors to eliminate processing by hand? One
approach would be to supply corrected info as a set of external patch
files, keyed to the RB version, that could be merged on the fly with
the output as needed.

(Feel free to respond off-list; I am just replying here so that others
can see there was activity.)

lj
_______________________________________________
Unsubscribe or switch delivery mode:
<http://www.realsoftware.com/support/listmanager/>

Search the archives of this list here:
<http://support.realsoftware.com/listarchives/lists.html>

Reply via email to