Added handling of structured content for edition values.

csv2cmi.py

@@ -15,7 +15,8 @@
 from csv import DictReader
 from datetime import datetime
 from os import path
-from xml.etree.ElementTree import Element, SubElement, Comment, ElementTree
+from xml.etree.ElementTree import (Element, SubElement, Comment, ElementTree,
+                                   fromstring, ParseError)
 
 __license__ = "MIT"
 __version__ = '2.0.1'
@@ -341,8 +342,28 @@ def createPlaceName(placeNameText, placeNameRef):
 
 def createEdition(biblText, biblType, biblID):
     """Create a new bibliographic entry."""
-    bibl = Element('bibl')
-    bibl.text = biblText
+    try:
+        bibl = fromstring(biblText)
+        if bibl.tag != 'bibl':
+            logging.warning('Structured edition "%s" has wrong root ' +
+                            'element <%s> given in csv file. Only ' +
+                            '<bibl>-element is allowed.',
+                            biblText, bibl.tag)
+            raise Exception
+    except ParseError as e:
+        bibl = Element('bibl')
+        bibl.text = biblText
+        # error code 2 is syntax error and this should be the case for all
+        # unstructured titles. Higher error codes are XML specific like
+        # completeness, well-formedness or mismatching tags etc. We assume,
+        # that those codes are produced by bad but wanted structures, so we
+        # leave a warning message.
+        if e.code > 2:
+            logging.warning('If edition value "%s" should be structured. ' +
+                            'It\'s not parsable. Error: %s', biblText, e)
+    except Exception:
+        bibl = Element('bibl')
+        bibl.text = biblText
     bibl.set('type', biblType)
     bibl.set('xml:id', biblID)
     return bibl
@@ -435,7 +456,8 @@ def processPlace(letter, correspondent):
 root = Element('TEI')
 root.set('xmlns', ns.get('tei'))
 root.append(
-    Comment(' Generated from table of letters with csv2cmi ' + __version__ + ' '))
+    Comment(' Generated from table of letters with csv2cmi ' +
+            __version__ + ' '))
 
 # teiHeader
 teiHeader = SubElement(root, 'teiHeader')
@@ -558,7 +580,7 @@ def processPlace(letter, correspondent):
             profileDesc.append(entry)
 
 # replace short titles if configured
-for bibl in sourceDesc.findall('bibl'):
+for idx, bibl in enumerate(sourceDesc.findall('bibl')):
     # Try to use bibliographic text as key for section in config file
     editionKey = bibl.text
     try:
@@ -568,16 +590,48 @@ def processPlace(letter, correspondent):
         except configparser.NoOptionError:
             # if type is not set, use the default one
             pass
-        bibl.text = editionTitle
-        bibl.set('type', editionType)
+        # Remember id to add it later, in case we got structured content.
+        id = bibl.get('xml:id')
+        try:
+            # Try to parse title as XML to allow structured content
+            bibl_new = fromstring(editionTitle)
+            if bibl_new.tag != 'bibl':
+                logging.warning('Structured title for key "%s" has wrong ' +
+                                'root element <%s> given in ini file. Only ' +
+                                '<bibl>-element is allowed.',
+                                editionKey, bibl_new.tag)
+                raise Exception
+            bibl_new.set('xml:id', id)
+            bibl_new.set('type', editionType)
+            # Setting bibl = bibl_new doesn't work. We have to explicitly
+            # replace the original bibl element by the new structured bibl
+            # element.
+            sourceDesc.remove(bibl)
+            sourceDesc.insert(idx, bibl_new)
+        except ParseError as e:
+            bibl.text = editionTitle
+            bibl.set('type', editionType)
+            # error code 2 is syntax error and this should be the case for all
+            # unstructured titles. Higher error codes are XML specific like
+            # completeness, well-formedness or mismatching tags etc. We assume,
+            # that those codes are produced by bad but wanted structures, so we
+            # leave a warning message.
+            if e.code > 2:
+                logging.warning('If title for key "%s" in ini file should ' +
+                                'be structured. It\'s not parsable. Error: %s',
+                                editionKey, e)
+        except Exception:
+            bibl.text = editionTitle
+            bibl.set('type', editionType)
     except configparser.NoOptionError:
         logging.warning(
-            'Incomplete section %s in ini file. Title and type option must be set.', editionKey)
+            'Incomplete section %s in ini file. ' +
+            'Title and type option must be set.', editionKey)
     except configparser.NoSectionError:
-        # if there is no matching section, we assume that there should be no one
+        # if there is no matching section, we assume that there should be
+        # no one.
         pass
 
-
 # generate empty body
 root.append(createTextstructure())
 

examples/shorttitle-replacement/Example.csv

@@ -2,3 +2,5 @@ sender,senderID,senderPlace,senderPlaceID,senderDate,senderDateText,addressee,ad
 Georges-Louis Le Sage,116948078,Genf,,1762-01-27,27. Januar 1762,Friedrich Heinrich Jacobi,118556312,,,,,JBWI1,2,
 Friedrich Heinrich Jacobi,118556312,,,1773-02-02,2. Februar 1773,Johann Georg Jacobi,118775782,,,,,JBWI4,*276.2,
 Johann Wolfgang Goethe,118540238,,,"[1774-09,1774-10]",September oder Oktober 1774,Friedrich Heinrich Jacobi,118556312,,,,,JBWI10,*349.2,
+Phillipp Jacob Spener,,Frankfurt a. M.,,1666-09-21,21. September 1666,Gottlieb Spizel,,Augsburg,,,,SPBW-DD-1,1,
+Phillipp Jacob Spener,,Frankfurt a. M.,,1666-11-01,1. November 1666,Balthasar Scheidt,,Straßburg,,,,"<bibl>Philipp Jakob Spener, Briefe aus der Dresdner Zeit 1686 - 1691, Band 1: 1686-1687, hrsg. von Johannes Wallmann in Zusammenarbeit mit Martin Friedrich, Klaus vom Orde und Peter Blastenbrei, Tübingen : Mohr Siebeck, 2003, ISBN 3-16-147427-9, <ref target=""http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa2-77749"">urn:nbn:de:bsz:14-qucosa2-77749</ref>.</bibl>",2,

examples/shorttitle-replacement/Example.xml

@@ -1,2 +1,2 @@
 <?xml version='1.0' encoding='utf-8'?>
-<TEI xmlns="http://www.tei-c.org/ns/1.0"><!-- Generated from table of letters with csv2cmi 2.0.0-beta --><teiHeader><fileDesc><titleStmt><title xml:id="title_4u2r1isd_vb14qyqn">Example file to demonstrate the replacement of shorttitles</title><editor>Uwe Kretschmer</editor></titleStmt><publicationStmt><publisher>Sächsische Akademie der Wissenschaften zu Leipzig</publisher><idno type="url">https://raw.githubusercontent.com/saw-leipzig/csv2cmi/master/example/shorttitle-replacement/Example.xml</idno><date when="2019-03-15T19:24:00.409619" /><availability><licence target="https://creativecommons.org/licenses/by/4.0/">This file is licensed under the terms of the Creative-Commons-License CC-BY 4.0</licence></availability></publicationStmt><sourceDesc><bibl type="online" xml:id="edition_ismw7t4y_2ve0ob2p">Friedrich Heinrich Jacobi, Band I.1: Briefwechsel 1762–1775, 1981. frommann-holzbog</bibl><bibl type="print" xml:id="edition_95ssee8d_i0m60lgc">Friedrich Heinrich Jacobi, Band I.4: Briefwechsel 1785. Nachtrag zum Briefwechsel 1764–1784. 2003, frommann-holzbog</bibl><bibl type="print" xml:id="edition_tyin9jcs_rbceg1n4">Friedrich Heinrich Jacobi, Band I.10: Briefwechsel Juni 1792 bis September 1794. Nachtrag zum Briefwechsel 1769–1789. 2015, frommann-holzbog</bibl></sourceDesc></fileDesc><profileDesc><correspDesc key="2" source="#edition_ismw7t4y_2ve0ob2p" xml:id="letter_16nlaemq_rh3p2814"><correspAction type="sent" xml:id="sender_l3ca2e5x_tuemr26n"><persName ref="http://d-nb.info/gnd/116948078">Georges-Louis Le Sage</persName><placeName>Genf</placeName><date when="1762-01-27">27. Januar 1762</date></correspAction><correspAction type="received" xml:id="addressee_w004vuqi_yujtan6d"><persName ref="http://d-nb.info/gnd/118556312">Friedrich Heinrich Jacobi</persName></correspAction></correspDesc><correspDesc key="*276.2" source="#edition_95ssee8d_i0m60lgc" xml:id="letter_55vke6aa_57er6r8e"><correspAction type="sent" xml:id="sender_9sjzsc0l_sd9ypb29"><persName ref="http://d-nb.info/gnd/118556312">Friedrich Heinrich Jacobi</persName><date when="1773-02-02">2. Februar 1773</date></correspAction><correspAction type="received" xml:id="addressee_ub0ykgtr_izjvyx0p"><persName ref="http://d-nb.info/gnd/118775782">Johann Georg Jacobi</persName></correspAction></correspDesc><correspDesc key="*349.2" source="#edition_tyin9jcs_rbceg1n4" xml:id="letter_ha9wsagx_qlh6tmua"><correspAction type="sent" xml:id="sender_0k692gos_1t4ajst5"><persName ref="http://d-nb.info/gnd/118540238">Johann Wolfgang Goethe</persName><date notAfter="1774-10" notBefore="1774-09">September oder Oktober 1774</date></correspAction><correspAction type="received" xml:id="addressee_pq1jdzb4_0uig88t0"><persName ref="http://d-nb.info/gnd/118556312">Friedrich Heinrich Jacobi</persName></correspAction></correspDesc></profileDesc></teiHeader><text><body><p /></body></text></TEI>
+<TEI xmlns="http://www.tei-c.org/ns/1.0"><!-- Generated from table of letters with csv2cmi 2.0.1 --><teiHeader><fileDesc><titleStmt><title xml:id="title_02cxyw59_n2ra3can">Example file to demonstrate the replacement of shorttitles</title><editor>Uwe Kretschmer</editor></titleStmt><publicationStmt><publisher>Sächsische Akademie der Wissenschaften zu Leipzig</publisher><idno type="url">https://raw.githubusercontent.com/saw-leipzig/csv2cmi/master/example/shorttitle-replacement/Example.xml</idno><date when="2019-04-11T16:39:54.501822" /><availability><licence target="https://creativecommons.org/licenses/by/4.0/">This file is licensed under the terms of the Creative-Commons-License CC-BY 4.0</licence></availability></publicationStmt><sourceDesc><bibl type="online" xml:id="edition_9i7pia7d_m8uukean">Friedrich Heinrich Jacobi, Band I.1: Briefwechsel 1762–1775, 1981. frommann-holzbog</bibl><bibl type="print" xml:id="edition_i4cf54xv_74qrd6bw">Friedrich Heinrich Jacobi, Band I.4: Briefwechsel 1785. Nachtrag zum Briefwechsel 1764–1784. 2003, frommann-holzbog</bibl><bibl type="print" xml:id="edition_6da1ctpa_jumqixmn">Friedrich Heinrich Jacobi, Band I.10: Briefwechsel Juni 1792 bis September 1794. Nachtrag zum Briefwechsel 1769–1789. 2015, frommann-holzbog</bibl><bibl type="hybrid" xml:id="edition_jxdl1u8t_kczysdhi">Philipp Jakob Spener, Briefe aus der Dresdner Zeit 1686 - 1691, Band 1: 1686-1687, hrsg. von Johannes Wallmann in Zusammenarbeit mit Martin Friedrich, Klaus vom Orde und Peter Blastenbrei, Tübingen : Mohr Siebeck, 2003, ISBN 3-16-147427-9, <ref target="http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa2-77749">urn:nbn:de:bsz:14-qucosa2-77749</ref>.</bibl><bibl type="print" xml:id="edition_9jqqyyqd_w3phlhcm">Philipp Jakob Spener, Briefe aus der Dresdner Zeit 1686 - 1691, Band 1: 1686-1687, hrsg. von Johannes Wallmann in Zusammenarbeit mit Martin Friedrich, Klaus vom Orde und Peter Blastenbrei, Tübingen : Mohr Siebeck, 2003, ISBN 3-16-147427-9, <ref target="http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa2-77749">urn:nbn:de:bsz:14-qucosa2-77749</ref>.</bibl></sourceDesc></fileDesc><profileDesc><correspDesc key="2" source="#edition_9i7pia7d_m8uukean" xml:id="letter_xnhu3z1z_fb6ng8d1"><correspAction type="sent" xml:id="sender_bkhihqej_eywknir6"><persName ref="http://d-nb.info/gnd/116948078">Georges-Louis Le Sage</persName><placeName>Genf</placeName><date when="1762-01-27">27. Januar 1762</date></correspAction><correspAction type="received" xml:id="addressee_13avrqxg_fs9biza6"><persName ref="http://d-nb.info/gnd/118556312">Friedrich Heinrich Jacobi</persName></correspAction></correspDesc><correspDesc key="*276.2" source="#edition_i4cf54xv_74qrd6bw" xml:id="letter_4vfa4nbw_pkcqnx7m"><correspAction type="sent" xml:id="sender_f6ds38fk_a1avpk6e"><persName ref="http://d-nb.info/gnd/118556312">Friedrich Heinrich Jacobi</persName><date when="1773-02-02">2. Februar 1773</date></correspAction><correspAction type="received" xml:id="addressee_ah7z45b0_w9z7v6pk"><persName ref="http://d-nb.info/gnd/118775782">Johann Georg Jacobi</persName></correspAction></correspDesc><correspDesc key="*349.2" source="#edition_6da1ctpa_jumqixmn" xml:id="letter_llazj6gf_2y88yv5y"><correspAction type="sent" xml:id="sender_lwf2e2nm_7rurrj3o"><persName ref="http://d-nb.info/gnd/118540238">Johann Wolfgang Goethe</persName><date notAfter="1774-10" notBefore="1774-09">September oder Oktober 1774</date></correspAction><correspAction type="received" xml:id="addressee_s88cxp7b_ipe9l8jk"><persName ref="http://d-nb.info/gnd/118556312">Friedrich Heinrich Jacobi</persName></correspAction></correspDesc><correspDesc key="1" source="#edition_jxdl1u8t_kczysdhi" xml:id="letter_36n7gf1v_l6qpulsy"><correspAction type="sent" xml:id="sender_umt4wyt0_qjowcmnk"><persName>Phillipp Jacob Spener</persName><placeName>Frankfurt a. M.</placeName><date when="1666-09-21">21. September 1666</date></correspAction><correspAction type="received" xml:id="addressee_5m3b4l70_29nc00mi"><persName>Gottlieb Spizel</persName><placeName>Augsburg</placeName></correspAction></correspDesc><correspDesc key="2" source="#edition_9jqqyyqd_w3phlhcm" xml:id="letter_ou00b2hv_pl3fnxht"><correspAction type="sent" xml:id="sender_z9f5h1kd_7gdyc0jy"><persName>Phillipp Jacob Spener</persName><placeName>Frankfurt a. M.</placeName><date when="1666-11-01">1. November 1666</date></correspAction><correspAction type="received" xml:id="addressee_773twqht_qjk6jyks"><persName>Balthasar Scheidt</persName><placeName>Straßburg</placeName></correspAction></correspDesc></profileDesc></teiHeader><text><body><p /></body></text></TEI>

examples/shorttitle-replacement/csv2cmi.ini

@@ -18,3 +18,10 @@ type = print
 [JBWI10]
 title = Friedrich Heinrich Jacobi, Band I.10: Briefwechsel Juni 1792 bis September 1794. Nachtrag zum Briefwechsel 1769–1789. 2015, frommann-holzbog
 # type is optional, defaults to print
+
+# Especially in case You have a online or hybrid edition You may want to set a link to an online ressource within the TEI bibl-statement.
+# To do so, You can just insert the complete and well-formed bibl XML structure like below. The attributes xml:id and type will be overwritten
+# by the program yo you don't need to add them here. Wrong example would be <bibl xml:id="myID123" type="online">Text</bibl>.
+[SPBW-DD-1]
+title = <bibl>Philipp Jakob Spener, Briefe aus der Dresdner Zeit 1686 - 1691, Band 1: 1686-1687, hrsg. von Johannes Wallmann in Zusammenarbeit mit Martin Friedrich, Klaus vom Orde und Peter Blastenbrei, Tübingen : Mohr Siebeck, 2003, ISBN 3-16-147427-9, <ref target="http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa2-77749">urn:nbn:de:bsz:14-qucosa2-77749</ref>.</bibl>
+type = hybrid

tests/tests.csv

@@ -1,4 +1,6 @@
-sender,senderID,senderPlace,senderPlaceID,senderDate,addressee,addresseeID,addresseePlace,addresseePlaceID,edition,key,note
-Donald Trump,http://id.loc.gov/authorities/names/n85387872,Washington,http://www.geonames.org/4140963,2016-04-01,Donald Duck,http://id.loc.gov/authorities/names/no2016036349,Entenhausen,http://d-nb.info/gnd/4203723-2,Letters of Love an Ludicrousness,,LOC with wrong addresseePlaceID
-Πλάτων,http://d-nb.info/gnd/118594893,Athens,http://www.geonames.org/264371,[..-0350],Imperium Romanum,http://d-nb.info/gnd/1115828762,Rom,http://www.geonames.org/3169070/rome.html,Complete edition of never written letters,89274,EDTF one of a set with negative year; wrong rdf:type in addressee
-Πλάτων,http://viaf.org/viaf/108159964,Athens,http://www.geonames.org/264371,"{-400,-390,-370}",𐎧𐏁𐏂,http://viaf.org/viaf/293793834,Hañgmatana,,Complete edition of never written letters,10342,missing entry in GeoNames; countries can‘t be addressed with VIAF
+"sender","senderID","senderPlace","senderPlaceID","senderDate","addressee","addresseeID","addresseePlace","addresseePlaceID","edition","key","note"
+"Donald Trump","http://id.loc.gov/authorities/names/n85387872","Washington","http://www.geonames.org/4140963","2016-04-01","Donald Duck","http://id.loc.gov/authorities/names/no2016036349","Entenhausen","http://d-nb.info/gnd/4203723-2","Letters of Love an Ludicrousness",,"LOC with wrong addresseePlaceID"
+"Πλάτων","http://d-nb.info/gnd/118594893","Athens","http://www.geonames.org/264371","[..-0350]","Imperium Romanum","http://d-nb.info/gnd/1115828762","Rom","http://www.geonames.org/3169070/rome.html","Complete edition of never written letters",89274,"EDTF one of a set with negative year; wrong rdf:type in addressee"
+"Πλάτων","http://viaf.org/viaf/108159964","Athens","http://www.geonames.org/264371","{-400,-390,-370}","𐎧𐏁𐏂","http://viaf.org/viaf/293793834","Hañgmatana",,"Complete edition of never written letters",10342,"missing entry in GeoNames; countries can‘t be addressed with VIAF"
+"sender",,,,"2019-04-11","addressee",,,,"<wrongRoot>Structured <ref target=""URI"">edition</ref> title.</wrongRoot>",10000,"Only <bibl>-element is allowed as root for structured content of edition."
+"sender",,,,"2019-04-11","addressee",,,,"<some>bad</xml>",10001,"No well-formed XML."