pw/pwg literary source display

[funderburkjim]

This issue is provided as part of explanation of current cdsl display of literary source markup for pw dictionary. The explanation also applies to pwg dictionary.
This explanation is in reference to this comment at issue 95.

The explanation is done via an example from headword akupyant.

from pw.txt

<L>171<pc>1003-1<k1>akupyant<k2>a/kupyant<e>100
{#a/kupyant#}¦ <lex>Adj.</lex> {%nicht aufwallend%} <ls>AV. 29,130,8.</ls>
<LEND>

current display result

relevant html

This html is generated by the basicdisplay.php

<span class="ls">
  <span style="color:blue;" class="ls">Av.</span>
  <span class="ls"> 29,130,8.</span>
</span>

css for ls class

from 'basic.css'

.ls {color:gray; font-size:11pt; font-family: charterindocapital;}

line from pw/pywork/pwauth/pwbib_input.txt

1020	AV.	Av.	ATHARVAVEDA, Ausg. von ROTH und WHITNEY (ROTH). 

[funderburkjim]

basicdisplay.php matches the xml element <ls>AV. 29,130,8.</ls> with the above line of pwbib_input.txt.
This matches 'AV.' with THE line of pwbib_input.txt using the second field (AV.)

Once this line is identified, basicdisplay constructs the html. Note that the html uses the third field (Av.).

In rendering the html, the browser displays the abbreviation using the 'ls' css class, which in particular uses
the charterindocapital font. This font displays lower-case letters as 'small capital' glyphs.

[funderburkjim]

other dictionaries

The display for other dictionaries (other than pw, pwg) is similar, but not quite the same.
For example, consider the headword anusPura in Grassman:
In gra.txt,

<L>542<pc>0061<k1>anusPura<k2>anusPura/
{@(anusphurá),@}¦ <ab>a.</ab>, {%schwirrend%} [<ab>v.</ab> sphur <ab n="mit">m.</ab> ánu].
<div n="TS">-ám śarám <ls>AV. 1,2,3</ls>.
<LEND>

The matching line of the ls tooltip file for grassman has just 2 fields.

AV.	Atharva Veda

basicdisplay generates almost the same html as for pw:

<span class="ls">
  <span style="color:blue;" class="ls">AV.</span>
  <span class="ls"> 1,2,3</span></span>

But notice that we are now displaying the text 'AV.', again with the charerindocapital font; so both the 'A' and the 'V' are capital-letter-sized.

[funderburkjim]

The fact that in pwg the 'V' in 'AV.' is displayed as a small capital letter could be considered an error in the 3rd field of the line in pwbib_input.txt. This could be corrected by changing that line to

1020	AV.	AV.	ATHARVAVEDA, Ausg. von ROTH und WHITNEY (ROTH). 

[Andhrabharati]

@funderburkjim

My observation is wrt the bib_input file content (in PW family) itself!

The pwk book has the resp. entries (AK, AV, ṚV, TS, VP, etc.) in FULL capital letters, not in small-caps. (Ak, Av, Ṛv, Ts, Vp etc.); and so does PWG.

[Andhrabharati]

The fact that in pwg the 'V' in 'AV.' is displayed as a small capital letter could be considered an error in the 3rd field of the line in pwbib_input.txt. This could be corrected by changing that line to

1020	AV.	AV.	ATHARVAVEDA, Ausg. von ROTH und WHITNEY (ROTH). 

Exactly; and, I had posted the PWG bib file 2 years back with such corrections (see my above post) which got ignored thus far.

[funderburkjim]

Should we have that 3rd field at all?

The print of PW does often (but not always) use large/small capital letters.
e.g.

And current cdsl displays mimic the print fairly well (except for the capitalization errors in pwbib_input.txt such as for AV)

But sometimes the print uses more normal letters for the ls abbreviation. Such as for 'Spr.'

So for cases like 'Spr.' , the cdsl display does NOT mimic the print.

[Andhrabharati]

The print of PW does often (but not always) use large/small capital letters.

But sometimes the print uses more normal letters for the ls abbreviation. Such as for 'Spr.'

There is a clear logic behind what the print has!

And, I had suggested even to go for simplistic ls-tooltips like in MW for PW family as well.

As I had earlier said in one of my posts-- "No need to make life complex (unnecessarily), when it is simple otherwise".

[funderburkjim]

So, you would be in favor of changing pwbib_input.txt to the form of Grassman?

i.e. dropping the 1st and 3rd fields. example:

AV.  ATHARVAVEDA, Ausg. von ROTH und WHITNEY (ROTH). 

[Andhrabharati]

If agreed upon, I can help preparing the bib_input files for PWG and pwk, as in every other CDSL work.

In fact, at the end of my current pwk (re-)working, I had planned to give out my resolved ls-entries, which I had kept on hold thus far (though prepared 2 years back itself).

Now, as I am looking further into pwk, finding more ls-entries and also some corrections to the earlier ones marked.
[Last time, I had spent just two days on the pwk; but stopped abruptly.]

[Andhrabharati]

So, you would be in favor of changing pwbib_input.txt to the form of Grassman?

i.e. dropping the 1st and 3rd fields. example:

AV.  ATHARVAVEDA, Ausg. von ROTH und WHITNEY (ROTH). 

Not just that, I would even be splitting some of them further depending on context; you may note that the print has clubbed two or more separable entries together under a single biblio entry at few places.

See as an example, the Ramayana entry in pwk that clubbled Schlegel, Gorresio and Bombay editions together.

[funderburkjim]

If agreed upon, I can help preparing the bib_input files for pwk,

Agreed.

A copy of the current pwbib_input.txt file is in the issue95/litsrc directory of as pwbib_input_0.txt.

When you provide a revision of pwbib_input_0.txt, I will need to correspondingly modify

• pywork code to create the sqlite version (pwbib.sqlite) of the tooltip file from the revised pwbib_input.txt
  • https://github.com/sanskrit-lexicon/csl-pywork/tree/master/v02/distinctfiles/pw/pywork/pwauth
• basicdisplay program to use the revised 2-field format of that sqlite file