Hi again;<br /><br />Our group receives XML files that contain eng data for multiple widgets inside 1 standard component. Like....<br /><br /> <pre><br /> <component><br /> <Widget><br /> <name>My Thing</name><br /> <phone>555-1212</phon><br /> <description>Very short</description><br /> </Widget><br /><br /> <Widget><br /> <name>Their Thing</name><br /> <phone>429-4242</phone><br /> <description>Also short</description><br /> </Widget><br /> ...<br /> </component><br /> </pre><br />Unfortunately, the 3rd party SW ("MakeWidgets") only allows engs to write 1 para of Widget description - no tables, graphics, etc - and our Widgets get complicated. We've chatted with the SW company about this restriction but alas, "MakeWidgets" can't change anytime soon.<br /><br />So our group is trying to split out Widgets separately, treat them as Frame read-only text insets, then interleave long description areas. Like....<br /><br /><pre><br /> <ModifiedComponent><br /> <Widget IsTextInset="Yes"><br /> <name>My Thing</name><br /> <phone>555-1212</phon><br /> <description>Very short</description><br /> </Widget><br /><br /> <LongDescription BelongsTo="My Thing"><br /> Blah, blah<p/><br /> <table/><br /> </LongDescription><br /><br /> <WidgetIsTextInset="Yes"><br /> <name>Their Thing</name><br /> <phone>429-4242</phone><br /> <description>Also short</description><br /> </Widget><br /> <br /> <LongDescription BelongsTo="Their Thing"><br /> Blah, blah<p/><br /> <GraphicHere/><br /> </LongDescription><br /> ...<br /><br /> </ModifiedComponent><br /></pre><br />The final system allows "MakeWidgets" to write Widget XML while Frame inputs Widget XML as<br />b read-only<br />text insets so Engs or TWs can see (but not write) the data while they create better descriptions. In the long run, we also want the "LongDescriptions" to be dbased text insets so we can reuse them in other docs.<br /><br />Last restriction. "MakeWidgets" can't see the "LongDescription" element so we need to fold everything back into a single file once we write the XML from inside Frame. This project entails 2 tasks (that I see):<br /><ul><br /><li>Break 1 component file into multiple text insets (or read 1 element<br />in the component file as 1 text inset, similar to a conref).</li><br /><br /><li>Interleave multiple files into 1 single file that Frame accepts as input. BTW, it's interesting that Frame is good at creating multiple output files, but not-so-good at accepting multiple files as input. But never mind.</li><br /></ul><br />Both tasks seem difficult to do, so I'm soliciting ideas and opinions. Also raucous laughter. ;)<br /><br />Any ideas?<br /><br />Thanks!<br /><br />---------------------------------------------------------------<br />Footnote. Yes, I know DITA can handle modules like this - but "MakeWidgets" doesn't accept DITA format. This is one reason why I don't like DITA - it's TW oriented and High Tech XML formats are eng driven. So either you don't use DITA or you swap formats - extremely hard to do - and track info in 2 places. Yuck!<br /><br />Sorry Ann Rockley.<br />---------------------------------------------------------------<br /><br />David Blyth (the UNIX and vi dinosaur)<br />Staff Technical Writer<br />QCT Division<br />QUALCOMM - Standard Disclaimers Apply<br /><br />Only 149,999,950 more years of Ruling The Earth to go<br />---------------------------------------------------------------
↧