Standardization of pwg links for KATHĀS.

[funderburkjim]

The 'normal' markup for PWG dictionary of a reference to Katha-sarit-sagara has the form <ls>KATHĀS. t, s.</ls> , where t is the taranga number and s is the shloka number (both are digit sequences). In a display for pwg, this form generates a link to the cdsl link target for Katha-sarit-sagara , as discussed in #70. Other activated link forms are <ls n="KATHĀS.">t, s.</ls> and <ls n="KATHĀS. t,">s.</ls>

But there are many 'implied' forms whose markup can be changed to a sequence of these normal forms. For instance, <ls>KATHĀS. 12, 33. 17, 17. 19, 50</ls> generates a link for 12,33 but not for 17,17 or 19,50.

This example can be recoded in pwg.txt as <ls>KATHĀS. 12, 33.</ls> <ls n="KATHĀS.">17, 17.</ls> <ls n="KATHĀS.">19, 50</ls>

In this recoding, 17,17 and 19,50 are active links. This issue describes work that recodes most of these implied forms.

[gasyoun]

This issue describes work that recodes most of these implied forms.

Work done?

[Andhrabharati]

Not done yet; when done, the KSS ls-links count would increase from 14k to 21k!!

[funderburkjim]

work finished

See issue71 files.

The readme file shows details.

Some stats:

• 2538 lines of pwg changed
• The changes were done in 3 steps:
  • change_1.txt 2167 lines changed (by program)
  • change_2.txt 146 lines changed (by program)
  • change_3.txt 230 lines changed (manual)
• start with 19502 individual KATHĀS ls refs (linksort1_0.txt)
• finish with 25005 references (linksort1_3.txt)
  • All but 15 of these are 'linkable'
  • 8 are parameter-less (<ls>KATHĀS.</ls>)
  • 7 have non-linkable parameters -- see linksort3_3.txt
    • one of these is in the pdf, but within an 'Einl.' of a taranga, rather than a shloka.

I think this issue can be closed now.