Insert appropriate nome name in relevant files

jcowey commented 5 years ago

The tag <placeName type="ancient" subtype="nome">U</placeName>

must be replaced with the relevant one. All that we have is the above tag. This is a result of the data we received.

Edelweiss commented 5 years ago

There are 6438 files containing this kind of information.

Examples

Pathyris https://github.com/DCLP/idp.data/blob/master/DCLP/1/555.xml#L46

Oxyrynchos https://github.com/DCLP/idp.data/blob/master/DCLP/61/60066.xml#L46

Tentyris https://github.com/DCLP/idp.data/blob/master/DCLP/99/98077.xml#L53

What should the placeholder »U« be replaced with?

jcowey commented 5 years ago

Would you please generate a distinct list of those placeNames which are siblings of <placeName type="ancient" subtype="nome">U</placeName> in a googlespreadsheet

jcowey commented 5 years ago

I generated a distinct list of <placeName type="ancient" ref="[^<]+">[^<]+</placeName> at 311, but many of those were not accompanied by <placeName type="ancient" subtype="nome">U</placeName>

jcowey commented 5 years ago

I would then fill in the relevant replacements for »U« On the basis of such a list you would then be able to change and replace in appropriate fashion.

Edelweiss commented 5 years ago

https://docs.google.com/spreadsheets/d/1uNsT53My4l0JdcE8jogBsqVZyYHjWsWuIw5Dh5GoCI0/edit?usp=sharing

jcowey commented 5 years ago

https://github.com/DCLP/idp.data/blob/master/DCLP/1/555.xml#L46-L48 now looks like

   <placeName type="ancient" subtype="nome">U</placeName>
   <placeName type="ancient" subtype="region">Egypt</placeName>
   <placeName type="ancient" ref="http://www.trismegistos.org/place/1628">Pathyris</placeName>

should look be

   <placeName type="ancient" subtype="nome" ref="http://www.trismegistos.org/place/2849">Pathyrites</placeName>
   <placeName type="ancient" subtype="region">Egypt</placeName>
   <placeName type="ancient" ref="http://www.trismegistos.org/place/1628">Pathyris</placeName>

The same should of course be the case for https://github.com/DCLP/idp.data/blob/master/DCLP/1/555.xml#L53-L55

https://github.com/DCLP/idp.data/blob/master/DCLP/1/555.xml#L41 now offers

<origPlace>Found: Pathyris (U, Egypt); written: Pathyris (U, Egypt)</origPlace>

It should be

<origPlace>Found: Pathyris (Pathyrites, Egypt); written: Pathyris (Pathyrites, Egypt)</origPlace>

jcowey commented 5 years ago

https://github.com/DCLP/idp.data/blob/master/DCLP/61/60066.xml#L46-L48 offers

<placeName type="ancient" subtype="nome">U</placeName>
<placeName type="ancient" subtype="region">Egypt</placeName>
<placeName type="ancient" ref="http://www.trismegistos.org/place/1524">Oxyrynchos</placeName>

It should be

<placeName type="ancient" subtype="nome" ref="http://www.trismegistos.org/place/2722">Oxyrhynchites</placeName>
<placeName type="ancient" subtype="region">Egypt</placeName>
<placeName type="ancient" ref="http://www.trismegistos.org/place/1524">Oxyrhynchos</placeName>

Please note "Oxyrynchos" => "Oxyrhynchos"

https://github.com/DCLP/idp.data/blob/master/DCLP/61/60066.xml#L41 offers

<origPlace>Found: Oxyrynchos (U, Egypt); written: Egypt</origPlace>

it should be

<origPlace>Found: Oxyrhynchos (Oxyrhynchites, Egypt); written: Egypt</origPlace>

jcowey commented 5 years ago

Please leave the purple ones

Edelweiss commented 5 years ago

Beispiele für verschiedene Schreib- und Fundorte

/DCLP/109/108739.xml

<history>
  <origin>
     <origPlace>Found: Apollonopolis (U, Egypt); written: Latopolis (U, Egypt)</origPlace>
     <origDate notBefore="0500" notAfter="0699">500 - 699</origDate>
  </origin>
  <provenance type="found">
     <p>
        <placeName type="ancient" subtype="nome">U</placeName>
        <placeName type="ancient" subtype="region">Egypt</placeName>
        <placeName type="ancient" ref="http://www.trismegistos.org/place/269">Apollonopolis</placeName>
     </p>
  </provenance>
  <provenance type="composed">
     <p>
        <placeName type="ancient" subtype="nome">U</placeName>
        <placeName type="ancient" subtype="region">Egypt</placeName>
        <placeName type="ancient" ref="http://www.trismegistos.org/place/1227">Latopolis</placeName>
     </p>
  </provenance>
</history>

Weitere Beispiele

Edelweiss commented 5 years ago

Typen

Found|written
Fundort|Schreibort
Schreibort|Fundort

Mapping auf die Datenelemente

Schreibort / written > composed
Fundort / found > found

Mögliche Kombinationen

found|composed
composed|found

Edelweiss commented 5 years ago

https://github.com/DCLP/idp.data/commit/346523e9150aec4a620b7a4e23b438662455c387 https://github.com/DCLP/idp.data/commit/1a64dda73ef2ee0d0771361596eea978630443e0

Beispiele:

https://github.com/DCLP/idp.data/blob/master/DCLP/65/64487.xml https://github.com/DCLP/idp.data/blob/master/DCLP/66/65099.xml https://github.com/DCLP/idp.data/blob/master/DCLP/66/65067.xml https://github.com/DCLP/idp.data/blob/master/DCLP/49/48854.xml

jcowey commented 5 years ago

Looks excellet thank you. Would you now please create another pass at the Sheet "X für ein U" in the spreadsheet DCLP Geo https://docs.google.com/spreadsheets/d/1uNsT53My4l0JdcE8jogBsqVZyYHjWsWuIw5Dh5GoCI0/edit#gid=0

For the remaining cases of the use of "u" which you did not change because I was unsure of which nome was to be used.

Then I can go ahead and do the necessary work. If you do a new read out it will probably simply corroborate the rows I have marked in purple, but it is probably worth doing a new read out.

jcowey commented 5 years ago

Thank you. I see that you have now marked rows 16, 18-19 and 21-23 as one which correspond the the purple I had in the previous version.

Are the ones you are marking with HTTPS ones you did not catch previously. I had been surprised to see them, because I thought they would have been dealt with anyway.

So a second read out was indeed a very useful check.

jcowey commented 5 years ago

Oh sorry, that was my fault. I had pasted a https URL and not cleaned it as I had done with the others

Edelweiss commented 5 years ago

Ja, absolut! Gute Idee… Danke! Uff, ich bin gerade am sondieren, warum manche aufgenommen wurden, von denen man es nicht erwartet hätte, und warum andere offenbar nicht gematcht wurden.

Edelweiss commented 5 years ago

Ja, das mit dem HTTP(S) ist wohl deswegen entstanden, weil die Daten in Spalte »place name« und »place ref« nur Beispiele waren. Das war etwas irreführend und mir später selbst nicht mehr bewusst.

Edelweiss commented 5 years ago

Frage: Sollen die refs alle einheitlich auf entweder HTTP oder HTTPS lauten? Als Vorab-Clean-Up.

jcowey commented 5 years ago

I think we should go for HTTPS. It is what the URL (e.g. www.trismegistos.org/place/1008) resolves to. And everything is moving towards HTTPS nowadays

jcowey commented 5 years ago

I feel sure it is not like that in papyri.info, but that would be easy to fix and is for another day.

Edelweiss commented 5 years ago

httpS

https://github.com/DCLP/idp.data/commit/1535ee172ae587fd01f32f98c6298775511c0d1f https://github.com/DCLP/idp.data/commit/f115e30a28c961eaef15156f2233f4f128d8301d

X for U https://github.com/DCLP/idp.data/commit/b10bffe80859b3672baf9b8688ef9247cc656e7d

X for U 2 https://github.com/DCLP/idp.data/commit/5f288c93fcb82998fcf6773c20b985a652edd801 https://github.com/DCLP/idp.data/commit/f59ecf977ff11ed627c1dbdd0a3c6681dbd121da https://github.com/DCLP/idp.data/commit/e9665c4e63d102ce505c4904da1ef3c1a5f48132

Nome https://github.com/DCLP/idp.data/commit/7b0f2dae08a4da4efaa0ce7ff07444adfdedf34c

Region https://github.com/DCLP/idp.data/commit/80e4def813f94ec2fb7444513ea11c2bb86c99fe

X for U 3 https://github.com/DCLP/idp.data/commit/88a60f7571a1344348b58580e5758f2a638dce11

Edelweiss commented 5 years ago

Nach dem heutigen commit gibt es nun keine U-Orte mehr.