From 0ddbde5403c9acc700ed1ded0b487e512472d75f Mon Sep 17 00:00:00 2001
From: Klaus Thoden <kthoden@mpiwg-berlin.mpg.de>
Date: Mon, 5 Mar 2018 16:22:47 +0100
Subject: [PATCH] Generate a report of errors

---
 prepare_tei.py | 60 ++++++++++++++++++++++++++++++++++++++------------
 1 file changed, 46 insertions(+), 14 deletions(-)

diff --git a/prepare_tei.py b/prepare_tei.py
index 55b9024..170ee1f 100644
--- a/prepare_tei.py
+++ b/prepare_tei.py
@@ -105,6 +105,8 @@ def convert_citations(string):
 def parse_cited_range(list_of_xml_elements):
     """citedRange: split up parameters or remove element if attributes are empty"""
 
+    unsplittable_pageref = []
+
     for reference in list_of_xml_elements:
         cited_range = reference.find("t:citedRange", namespaces=NS_MAP)
         from_value = (cited_range.get("from"))
@@ -123,22 +125,32 @@ def parse_cited_range(list_of_xml_elements):
             cited_range.set("to", split_values[2])
         else:
             logging.info("Splitting the page range produced unexpected result. Tried to split %s" % from_value)
+            unsplittable_pageref.append(from_value)
+
+    return unsplittable_pageref
 # def parse_cited_range ends here
 
 def validate_citations(used_citekeys, bibdata):
-    """Check if all found citekeys are in the database"""
+    """Check if all found citekeys are in the database
+
+    Return a list of unavailable citekeys."""
 
     available_citekeys = bibdata.keys()
 
+    no_citekey = []
+
     for citekey in used_citekeys:
         if citekey not in available_citekeys:
+            no_citekey.append(citekey)
             logging.info("%s is not in the bibliographic database" % citekey)
+
+    return no_citekey
 # def validate_citations ends here
 
 def convert_figures(string):
     """Find figures shorthands"""
 
-    graphic_pattern = re.compile(r"\+(.*?)\+")
+    graphic_pattern = re.compile(r"(?<!\\)\+(.*?)\+")
 
     # +Fig.1CarteDuCielPotsdam!Glass photographic plate from the Carte
     # du Ciel survey, Potsdam Observatory, Plate 5, taken January 11,
@@ -158,6 +170,15 @@ def convert_figures(string):
 def make_figure_elements(list_of_figures, figure_directory):
     """Construct the figure element."""
 
+    bad_images = []
+    available_images = []
+    available_images_long = os.listdir(figure_directory)
+
+    for img in available_images_long:
+        available_images.append(os.path.splitext(img)[0])
+
+    print(available_images)
+
     for figure in list_of_figures:
         parent_tag = figure.getparent()
         parent_tag.tag = "figure"
@@ -167,23 +188,27 @@ def make_figure_elements(list_of_figures, figure_directory):
 
         parts = original_string.split("!")
 
-        if len(parts) == 3:
-            logging.info("This figure contains hyperimage directions")
-        elif len(parts) == 2:
-            available_images = os.listdir(figure_directory)
+        if len(parts) in range(2,4):
+            if parts[0] in available_images or parts[0] in available_images_long:
+                selected_image = parts[0]
+                logging.info("Found %s in the text. Selected %s as corresponding image." % (parts[0], selected_image))
+                figure.set("url", "images/" + selected_image)
+            else:
+                bad_images.append(original_string)
 
-            for image in available_images:
-                if image.startswith(parts[0]):
-                    selected_image = image
-
-            logging.info("Found %s in the text. Selected %s as corresponding image." % (parts[0], selected_image))
-
-            figure.set("url", "images/" + selected_image)
             head_element = etree.Element("head")
             parent_tag.insert(1, head_element)
             head_element.text = parts[1]
+
+            if len(parts) == 3:
+                logging.info("This figure contains hyperimage directions")
+                yenda_command = etree.Comment("Hyperimage direction: %s" % parts[2])
+                parent_tag.append(yenda_command)
+
         else:
-            logging.info("The figure string could be be split by '!': %s" % figure)
+            logging.info("The figure string could not be split by '!': %s" % figure)
+
+    return bad_images
 # def make_figure_elements ends here
 
 def cleanup_xml(xml_tree):
@@ -279,6 +304,12 @@ def fix_tei_header(xml_tree):
     return xml_tree
 # def fix_tei_header ends here
 
+def evaluate_report(report):
+    """Print collection of not found items"""
+
+    print(report)
+# def evaluate_report ends here
+
 def main():
     """The main bit"""
 
@@ -374,6 +405,7 @@ def main():
     tree.write(output, pretty_print=True, xml_declaration=True,encoding="utf-8")
     logging.info("Wrote %s." % output)
 
+    evaluate_report(report)
 # def main ends here
 
 if __name__ == '__main__':