Malformed dclp-hybrid values

Question

Malformed dclp-hybrid values

hcayless opened this issue 7 years ago · comments

There are 233 DCLP documents that have broken dclp-hybrid <idno> values. A full list can be found at https://gist.github.com/hcayless/0c99cb6af2b27239f397ca854e52e677. They all seem to be P.Herc. docs.

This error prevents correct indexing of the documents for search.

Tom Elliott commented 7 years ago

@jcowey ?

James M.S. Cowey · Answer 1 · Thu Aug 17 2017 22:46:20 GMT+0800 (China Standard Time)

@HolgerEssler this needs to be sorted as soon as we can manage. Do you want to have all P.Herc. publications collected under one standardised dclp-hybrid so that they ressemble e.g. p.oxy;12;1234?
If the answer is yes then we have to change
<idno type="dclp-hybrid">P.Herc. 1120</idno>
into
<idno type="dclp-hybrid">p.herc;;1120</idno> (that assumes no volume)

If the answer is no, then we replace these dclp-hybrid with the relevant "na;;23456" value, that is "na" (viz. no author) followed by the TM number.

Holger Essler · Answer 2 · Thu Aug 17 2017 23:47:06 GMT+0800 (China Standard Time)

Yes, please change
<idno type="dclp-hybrid">P.Herc. 1120</idno>
into
<idno type="dclp-hybrid">p.herc;;1120</idno>.
I suppose
<idno type="dclp-hybrid">P.Herc. 1043 + 1045</idno>
should then become
<idno type="dclp-hybrid">p.herc;;1043;1045</idno>
and
<idno type="dclp-hybrid">P.Herc. 419, 697, 1634</idno>
should become
<idno type="dclp-hybrid">p.herc;;419;697;1634</idno>.
Would that be ok?

Hugh A. Cayless · Answer 3 · Thu Aug 17 2017 23:58:07 GMT+0800 (China Standard Time)

I would recommend something like:

<idno type="dclp-hybrid">P.Herc. 1043 + 1045</idno> -> <idno type="dclp-hybrid">p.herc;;1043+1045</idno>
and
<idno type="dclp-hybrid">P.Herc. 419, 697, 1634</idno> -> <idno type="dclp-hybrid">p.herc;;419,697,1634</idno>

Tom Elliott · Answer 4 · Fri Aug 18 2017 06:28:50 GMT+0800 (China Standard Time)

@jcowey can this be done in Heidelberg?

On Thu, Aug 17, 2017 at 10:58 AM Hugh A. Cayless ***@***.***> wrote: I would recommend something like: <idno type="dclp-hybrid">P.Herc. 1043 + 1045</idno> -> <idno type="dclp-hybrid">p.herc;;1043+1045</idno> and <idno type="dclp-hybrid">P.Herc. 419, 697, 1634</idno> -> <idno type="dclp-hybrid">p.herc;;419,697,1634</idno> — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#317 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAQEdfx_I6noYANOuw16ihBHGNdz9ckKks5sZGMPgaJpZM4O6V6p> .

-- -- Tom Elliott, Ph.D. Associate Director for Digital Programs and Senior Research Scholar Institute for the Study of the Ancient World (NYU) http://isaw.nyu.edu/people/staff/tom-elliott Humanities Commons: @paregorios <https://hcommons.org/members/paregorios/> OrcID: 0000-0002-4114-6677 <http://orcid.org/0000-0002-4114-6677>

Hugh A. Cayless · Answer 5 · Wed Aug 23 2017 20:42:51 GMT+0800 (China Standard Time)

Possibly this warrants a new ticket, but a number of DCLP records have other problems in the dclp-hybrid idno, namely characters that cause problems in processing. See the list below:

James M.S. Cowey · Answer 6 · Wed Aug 23 2017 20:54:02 GMT+0800 (China Standard Time)

pretty sure I know how to fix this and will do so

o.frangé;;438 => o.frange;;438; as in ddbdp
o.wångstedt;;80 => o.wangstedt;;80; will have to be added to collection.rdf
p.genève[horssérie];;1 => p.geneve[horsserie];;1; will have to be added to collection.rdf
p.murabba'ât;2;108 => p.mur;2;108; as in ddbdp
p.demarée;;5 => p.demaree;;5; will have to be added to collection.rdf

is that analysis correct @hcayless ?

Hugh A. Cayless · Answer 7 · Wed Aug 23 2017 20:59:36 GMT+0800 (China Standard Time)

I'm not sure how the square brackets will play. We'll have to see.

James M.S. Cowey · Answer 8 · Wed Aug 23 2017 21:08:14 GMT+0800 (China Standard Time)

So have now created a new issue #324, to keep these two separate.

James M.S. Cowey · Answer 9 · Wed Aug 23 2017 21:54:43 GMT+0800 (China Standard Time)

@hcayless would you please check that
<idno type="dclp-hybrid">p.herc;;.+</idno>
is now fine.
There should now be no more
<idno type="dclp-hybrid">P.Herc.
left in
https://github.com/DCLP/idp.data/tree/master/DCLP

I have made a number of commits to make the required corrections.

Elemmire · Answer 10 · Wed Feb 28 2018 23:19:25 GMT+0800 (China Standard Time)

files that still need repair

./63/62411.xml: P.Herc. 228, 403, 407, 1425, 1581
./63/62425.xml: P.Herc. 495
./63/62426.xml: P.Herc. 558
./63/62476.xml: P.Herc. 1471

change to…

./63/62411.xml: p.herc;;228,403,407,1425,1581
./63/62425.xml: p.herc;;495
./63/62426.xml: p.herc;;558
./63/62476.xml: p.herc;;1471

Elemmire · Answer 11 · Wed Mar 07 2018 21:56:49 GMT+0800 (China Standard Time)

case-sensitive search for P.Herc. in xpath tei:idno[@type='dclp-hybrid'] didn’t bring forth any further idnos of the kind

Elemmire · Answer 12 · Wed Mar 07 2018 22:13:12 GMT+0800 (China Standard Time)

https://github.com/DCLP/idp.data/tree/issue317

(in development and master)

Elemmire · Answer 13 · Thu Mar 08 2018 17:19:29 GMT+0800 (China Standard Time)

Files can be viewed on github and will be picked up with the next sync.