BetaMasaheft / Documentation

Die Schriftkultur des christlichen Äthiopiens: Eine multimediale Forschungsumgebung
3 stars 3 forks source link

catalogue PDF issues #2541

Open thea-m opened 6 months ago

thea-m commented 6 months ago

List not yet completed, publishing now just to save :) https://docs.google.com/spreadsheets/d/1GiiNvi5VlaPZnsG1u0_b5P7jOjtRXaYlCOJ-SH0atxo/edit#gid=0

Important

General

Decoration

Binding

Colophon

Quire structure

Condition

eu-genia commented 6 months ago

@thea-m you must always exactly show

what you have vs what you want to have

so far I see some vague wishes but I do not see precisely what is wrong and how exactly it should be corrected

eu-genia commented 6 months ago

I also do not see in the branch anything from your current catalogue version. Please update the branch you want me to work with with ALL the files so that I do not have to recreate them (drive, fo, etc). Thank you

eu-genia commented 6 months ago

the branch seems to lack also the content for /data/bibliography.xml' can it be? please make sure that the branch you want me to work with contains EVERYTHING that you are working with

eu-genia commented 6 months ago

I have now created the bibliography file but it would have been easier to simply share the same set of files. Make sure that what is on catalogue branch, except for the mss, is exactly what you are using

eu-genia commented 6 months ago

I think the first point https://github.com/BetaMasaheft/makepdf/issues/14 was not resolved already because you reacted with thumbs up to a selection of proposals, of which the last one does not require any changes in the code. So it was not clear that you want anything to be done here.

eu-genia commented 6 months ago

Your second point Content of page header of PDF: left page: title of catalogue (this is no problem); right page: MS Shelfmark (currently: left authors, right "catalogue"). A solution to have the running shelfmark is available in FO. is it not the one I had already resolved in https://github.com/BetaMasaheft/makepdf/issues/10 ?

eu-genia commented 6 months ago

https://github.com/BetaMasaheft/makepdf/issues/14 - this first issue can only be attempted to be solved if all the encoding is fully consistent.

e.g. if I tell the code to print the labels instead of the titles, it may work, but then you will have no titles in cases where there are no labels provided for textparts in works

<div type="textpart" subtype="chapter" xml:id="Monday"/>
            <div type="textpart" subtype="chapter" xml:id="Tuesday"/>
            <div type="textpart" subtype="chapter" xml:id="Wednesday"/>
            <div type="textpart" subtype="chapter" xml:id="Thursday"/>
            <div type="textpart" subtype="chapter" xml:id="Friday"/>
            <div type="textpart" subtype="chapter" xml:id="Saturday"/>
            <div type="textpart" subtype="chapter" xml:id="Zekra">

Pietro suggested two options

  1. the function can be replicated in the PDF.xql script to behave differently, pointing thus to a local function rather than to the API
  2. editing the XML to contain the desired title as text only for that specific occurrence.

In any case change in data is necessary. Option 2 does not request any changes in the code.

thea-m commented 6 months ago

Thank you! I have updated the branch with the files.

eu-genia commented 6 months ago

I have fixed the labels (labels when present, IDs when not) and I have reinserted the change for headers.

For all points : please do not describe, show precisely (snapshots) what is wrong and how it should be. Thank you.

eu-genia commented 6 months ago

note that the code does not help when the problem is in the data. (1) for the moment only titles with xml:lang attribute are printed, so if there is no xml:lang attribute these will be ignored (2) for the moment the code looks in the works for the first listed Ethiopic title and prints it with its English translation, even if it comes as t3 or t4. (3) for titles in Fidel, the code looks for English first, so if the title is only in Fidel, there will be an empty space

this must be all clearly defined or there will be no good result - it must rely on very standardized data or have a very flexible and clearly defined code

eu-genia commented 6 months ago

I have relaxed the code and updated some of the work files, if errors arise the xml data should be fixed first (NB in some cases I have discovered the data is fixed in github but not on the app)

thea-m commented 6 months ago

The latest change in XQM now in the branch is with decorations - was that you? does it work? it is very difficult to work on the same files parallelly. If I should be doing anything noone else should (obviously Pietro already took the version without my updates, and now I see that the version of today is different from the one I had last week from you, this cannot work like this, I am sorry)

Thank you! The latest change from me was copying the files from my folder to the branch, so that this becomes authoritative. I am not doing any work in it. The decorations edit does not work (it is there for archival purposes as it used to work in PDFold.xql, but I must have made an error in copy pasting. that's why it's commented out)

thea-m commented 6 months ago

The titles are great like that, thank you. This is already a huge help! I will add the screenshots. For the general workflow, maybe we can discuss it orally tomorrow or in the next days rather than here.

eu-genia commented 6 months ago

Please clearly illustrate what you see and what you need instead. E.g.

eu-genia commented 6 months ago

pb/cb fixed https://github.com/BetaMasaheft/makepdf/commit/fb65dc2ceda8d0e315e12a5ee36cdbdba85efbc8 image

eu-genia commented 6 months ago
eu-genia commented 6 months ago

collation does not seem to match the XML, notes for items are not printed (intended?), numbering changed (intended?), loci for stubs etc not printed, image

<list>
                                            <item xml:id="q1" n="1">
                                                <dim unit="leaf">2</dim>
                                                <locus from="1r" to="2v"/>
                                            </item>
                                            <item xml:id="q2" n="2">
                                                <dim unit="leaf">8</dim>
                                                <locus from="3r" to="10v"/>
                                            </item>
                                            <item xml:id="q3" n="3">
                                                <dim unit="leaf">6</dim>
                                                <locus from="11r" to="16v"/>
                                                <note>The folios have been sewn together at a later stage.</note>
                                            </item>
                                            <item xml:id="q4" n="4">
                                                <dim unit="leaf">8</dim>
                                                <locus from="17r" to="24v"/>
                                            </item>
                                            <item xml:id="q5" n="5">
                                                <dim unit="leaf">6</dim>
                                                <locus from="25r" to="30v"/>
                                            </item>
                                            <item xml:id="q6" n="6">
                                                <dim unit="leaf">5</dim>
                                                <locus from="31r" to="35v"/>
                                            </item>
                                            <item xml:id="q7" n="7">
                                                <dim unit="leaf">8</dim>
                                                <locus from="36r" to="44v"/>
                                                stub between <locus target="#38"/> and <locus target="#39"/>
                                                stub between <locus target="#41"/> and <locus target="#42"/>
                                            </item>
                                            <item xml:id="q8" n="8">
                                                <dim unit="leaf">6</dim>
                                                <locus from="45r" to="50v"/>
                                            </item>
                                            <item xml:id="q9" n="9">
                                                <dim unit="leaf">3</dim>
                                                <locus from="51r" to="53v"/>
                                                <note>The folios have been stitched together at a later stage unto <ref target="#q10"/>.</note>
                                            </item>
                                            <item xml:id="q10" n="10">
                                                <dim unit="leaf">9</dim>
                                                <locus from="54r" to="62v"/>
                                                2 stubs between <locus target="#61"/> and <locus target="#62"/>
                                            </item>
                                            <item xml:id="q11" n="11">
                                                <dim unit="leaf">4</dim>
                                                <locus from="63r" to="67v"/>
                                            </item>
                                            <item xml:id="q12" n="12">
                                                <dim unit="leaf">9</dim>
                                                <locus from="68r" to="76v"/>
                                            </item>
                                            <item xml:id="q13" n="13">
                                                <dim unit="leaf">8</dim>
                                                <locus from="77r" to="84v"/>
                                            </item>
                                            <item xml:id="q14" n="14">
                                                <dim unit="leaf">8</dim>
                                                <locus from="85r" to="92v"/>
                                                3, stub after 6
                                                6, stub after 3
                                            </item>
                                            <item xml:id="q15" n="15">
                                                <dim unit="leaf">9</dim>
                                                <locus from="93r" to="101v"/>
                                            </item>
                                            <item xml:id="q16" n="16">
                                                <dim unit="leaf">7</dim>
                                                <locus from="102r" to="108v"/>
                                            </item>
                                            <item xml:id="q17" n="17">
                                                <dim unit="leaf">6</dim>
                                                <locus from="109r" to="114v"/>
                                                stub between <locus target="#113"/> and <locus target="#114"/>
                                            </item>
                                            <item xml:id="q18" n="18">
                                                <dim unit="leaf">9</dim>
                                                <locus from="115r" to="123v"/>
                                            </item>
                                        </list>

Does anything need to be fixed here? Would be good to closely compare the XML and the PDF, line by line, and mark what is missing and what is needed in this way. The bar at the end really seems to be the least relevant issue here.

Here there are several issues:

eu-genia commented 6 months ago
eu-genia commented 6 months ago

the collation code when describing the quire content (with s.l., stubs etc) has remained the same as in the app and works but the data is not clean the code expects things like (aeth c 14)

<item xml:id="q14" n="14">
                                                <dim unit="leaf">8</dim>
                                                <locus from="85r" to="92v"/>
                                                3, stub after 6
                                                6, stub after 3
                                            </item>

with folio numbers RELATIVE to the quire (this is also what is in the guidelines and expected from the main Bm app)

It cannot deal (but neither can the main Bm ap) with descriptions such as

<item xml:id="q10" n="10">
                                                <dim unit="leaf">9</dim>
                                                <locus from="54r" to="62v"/>
                                                2 stubs between <locus target="#61"/> and <locus target="#62"/>
                                            </item>

The collations should be corrected.

For now I have suppressed printing any text in the PDF.xql, but this does not resolve the actual problem of data incosistency, to be dealt with at some point.

image

(Also note that as a rule in Bm we do not split collations, we provide one collation for the entire codex, not for msParts)

eu-genia commented 6 months ago

decoNotes in decorations image NB the space after "harp" comes from the XML data, to fix there

eu-genia commented 6 months ago

NB the script does a lot on the BM server, causing java storage overflow when bigger collections of mss are processed.

At some point we should try to revise the pdf app to call at least some of the data locally or from GitHub. (Not so easy as the addresses vary but at least when there are lists)

eu-genia commented 6 months ago

There are still some open issues under makePDF where there is no comment as to what is lacking or I do not know why they are open as they are not repeated here, like here https://github.com/BetaMasaheft/makepdf/issues/18

eu-genia commented 6 months ago
eu-genia commented 6 months ago

@thea-m the error is not in the xql but in the XML data which has <origDate from="1500" to="1600" evidence="lettering"></origDate>, expected is @notBefore and @notAfter.

fixing the data gets image comma removed by https://github.com/BetaMasaheft/makepdf/commit/a7a3c7ddd4225e9dbe28e5c97d76e18b3ad1515e

eu-genia commented 6 months ago
thea-m commented 6 months ago

image

it should be printed as previously:

image IT WOULD BE PRINTED IF THE DATA WERE CONSISTENT, WITHOUT XML:LANG

The layout of Bibliography can be identical to that in composite manuscripts: image https://github.com/BetaMasaheft/makepdf/commit/0d3854ccf7ab26ef904d9fbe0fb3b37969c1fab6

eu-genia commented 6 months ago

the title is not printed because the code specifies <fo:block text-align="center" space-before="2mm" space-after="3mm">{$r//tei:titleStmt/tei:title[not(@xml:lang)]/text()}</fo:block>,

meaning "print title without xml:lang", clearly done on purpose (probably to exclude Arabic for DS) your data has <title xml:lang="en" xml:id="title1"><hi rendition="simple:italic">Malkǝʾ</hi>-hymns, Excerpts from the New Testament, Miracles of Mary, Miracles of Jesus, Homily by Jacob of Serug on the Angel and the Thief</title>

I suggest you remove the xml:lang="en" in manuscript titles

eu-genia commented 6 months ago