From b8aa5f5d9eee81b5f7a481f8779ae63f67878e4d Mon Sep 17 00:00:00 2001
From: Klaus Thoden <kthoden@mpiwg-berlin.mpg.de>
Date: Mon, 3 Dec 2018 16:15:07 +0100
Subject: [PATCH] Hyperimage integration

---
 tei2imxml.py | 99 +++++++++++++++++++++++++++++++++++++++++++---------
 1 file changed, 83 insertions(+), 16 deletions(-)

diff --git a/tei2imxml.py b/tei2imxml.py
index d09a1d5..29dfd6b 100644
--- a/tei2imxml.py
+++ b/tei2imxml.py
@@ -46,6 +46,8 @@
 
 CSL_FILE = CONFIG['Auxiliaries']['CSL_FILE']
 
+CSV_FILE = os.path.expanduser("hi_figures.csv")
+
 def get_publication_info(xml_tree):
     """Query the TEI document for metadata fields.
 
@@ -349,9 +351,64 @@ def format_authors(list_author_id, publang, xml_tree):
     return author_string
 # def format_authors ends here
 
-def transform_body(xml_tree, cited_data, publang):
+def hi_lookup(hi_id):
+    """Get hyperimage code from CSV file"""
+
+    import csv
+
+    nd = {}
+
+    logging.debug("Opening %s", CSV_FILE)
+
+    with open(CSV_FILE, newline='') as f:
+        reader = csv.DictReader(f, fieldnames = ("checked","id","number","elementstring"))
+        jsonStr = json.dumps(list(reader)[1:])
+        jsonObj = json.loads(jsonStr)
+
+        for xx in jsonObj[1:]:
+            nd[xx["id"]] = {"number": xx["number"], "checked" : xx["checked"], "elementstring" : xx["elementstring"]}
+
+    if nd[hi_id]:
+        hi_code = nd[hi_id]["elementstring"]
+    else:
+        logging.error("Could not find hi code %s", hi_id)
+
+    return hi_code
+# def hi_lookup ends here
+
+def transform_body(xml_tree, cited_data, publang, hyperimage=False):
     """Transform the body of XML document into IntermediateXML file"""
 
+    def handle_refs_default(ref):
+        """Handle refs the normal way"""
+
+        target_attribute = ref.get("target")
+        if not target_attribute:
+            logging.error("Found a ref element without target. Exiting.")
+            sys.exit()
+        else:
+            url_attribute = ref.get("type")
+            if url_attribute == "url":
+                del ref.attrib["type"]
+                del ref.attrib["target"]
+                ref.tag = "xref"
+                ref.set("url", target_attribute)
+            else:
+                ref.tag = "EOAref"
+                del ref.attrib["target"]
+                etree.SubElement(ref, "ref", teitarget=target_attribute)
+                etree.SubElement(ref, "Label").text = target_attribute
+        return
+    # def handle_refs_default ends here
+
+    def handle_refs_hyperimage(ref):
+        """Treat also the special cases of hyperimage refs"""
+
+        logging.info("Found a ref without target, what else have we got?")
+
+        return
+    # def handle_refs_hyperimage ends here
+
     logging.info("Performing XML transformations of the body.")
     ######################
     # Document structure #
@@ -516,6 +573,13 @@ def transform_body(xml_tree, cited_data, publang):
   </p>
 </EOAfigure>
 
+hyperimage
+<figure xml:id="chap14_fig2" corresp="#chap14_fig2-hi">
+
+<EOAfigure file="imagesTrnkova_CAS_572.jpg" hielement="ewkJfQoJCV0KCX0KfQ==" width="60px;" order="8" number="14.2">
+<caption>The town hall in Prague’s Academy of Sciences</caption>
+</EOAfigure>
+
     """
 
     figure_counter = 1
@@ -549,6 +613,15 @@ def transform_body(xml_tree, cited_data, publang):
                 figure_width = etree.SubElement(fig_p_element, "width").text = "60" #whatever
 
                 # <EOAfigurenonumber><p><file>images/1.jpg</file><width>33</width></p>
+            if figure_type == "hiviewer":
+                # display image in hyperimage viewer, not in lightbox
+                # hi_id needs to be looked up in hi_figures.csv
+                logging.debug("Found figure for hiviewer.")
+                hi_id = figure.get("corresp")[1:]
+                hi_code = hi_lookup(hi_id)
+                figure.set("hielement", hi_code)
+            else:
+                pass
 
             etree.strip_elements(figure, "{%s}graphic" % ns_tei)
 
@@ -631,22 +704,11 @@ def transform_body(xml_tree, cited_data, publang):
         ref_parent = ref.getparent()
         if ref_parent == "bibl":
             continue
-        target_attribute = ref.get("target")
-        if len(target_attribute) == 0:
-            print("Found a ref element without target. Exiting.")
-            sys.exit()
         else:
-            url_attribute = ref.get("type")
-            if url_attribute == "url":
-                del ref.attrib["type"]
-                del ref.attrib["target"]
-                ref.tag = "xref"
-                ref.set("url", target_attribute)
+            if hyperimage:
+                handle_refs_hyperimage(ref)
             else:
-                ref.tag = "EOAref"
-                del ref.attrib["target"]
-                etree.SubElement(ref, "ref", teitarget=target_attribute)
-                etree.SubElement(ref, "Label").text = target_attribute
+                handle_refs_default(ref)
 
     return xml_tree
 # def transform_body ends here
@@ -923,6 +985,7 @@ def main():
 
     parser = argparse.ArgumentParser()
     parser.add_argument("-d", "--pickleddata", default=data_pickle, help="Pickled data file to be used.")
+    parser.add_argument("-him", "--hyperimage", action="store_true")
     parser.add_argument("teifile", help="TEI XML file to convert into DocBook XML.")
     args = parser.parse_args()
 
@@ -999,7 +1062,11 @@ def main():
             formatted_references_dict[tmp_dict_key] = refs_for_bib_chapter
 
     tei_body = xml_tree.xpath("//t:body", namespaces=NS_MAP)[0]
-    body_transformed_tmp = transform_body(tei_body, cited_dict, publang=publication_language)
+    if args.hyperimage:
+        logging.info("Transforming body with Hyperimage support")
+    else:
+        pass
+    body_transformed_tmp = transform_body(tei_body, cited_dict, publang=publication_language, hyperimage=args.hyperimage)
     libeoaconvert.debug_xml_here(body_transformed_tmp, "body_transformed")
     body_transformed = etree.ElementTree(body_transformed_tmp)