risparser: fix handling of continuation lines

gargantext/util/parsers/RIS.py

@@ -47,11 +47,19 @@ class RISParser(Parser):
         for line in file:
             # bytes ~~> str
             line = line.decode("UTF-8").rstrip('\r\n')
+
+            # print("RIS line:", line)
+
             if len(line) >= 2 :
-                # extract the parameter key...
+                # print("(nonemptyline)")
+
+                # test if key line (otherwise: continuation line)
+                if match(r'[A-Z][A-Z0-9]\s', line):
                     parameter_key = line[:2]
+                    # print("(matchparamline:"+parameter_key+")")
 
-                # ...and keep the rest for when we know what to do with it
+                    # we can now be sure that the value is rest of the line
+                    # (keep it for when we know what to do with it)
                     current_value = line[self._begin:]
 
                     # it's a new key => therefore the previous key is finished
@@ -68,6 +76,7 @@ class RISParser(Parser):
                                     hyperdata[parameter["key"]] = final_value
                                 else:
                                     hyperdata = PY_values_decompose_and_save(final_value, hyperdata)
+                                # print("{saved previous"+last_key+"}")
 
                             #... or even finish the record (rare here, most often after empty line)
                             elif parameter["type"] == "delimiter":
@@ -76,6 +85,7 @@ class RISParser(Parser):
                                         if 'language_iso2' not in hyperdata.keys():
                                             hyperdata['language_iso2'] = 'en'
                                 yield hyperdata
+                                # print("{saved previous record}")
                                 last_key = None
                                 hyperdata = {}
 
@@ -83,13 +93,20 @@ class RISParser(Parser):
                         last_values = []
                         last_key = parameter_key
 
-                # 3 - new key or old: in any case we pass contents to
+                # continuation line: values start from position 0
+                else:
+                    current_value = line
+                    # print("(continuationline)")
+
+
+                # 3 - new key or old or no key: in any case we pass contents to
                 #     the value array buffer (=> for the next loop only)
                 last_values.append(current_value)
                 current_value = None
 
             # empty line => we need to check if PREVIOUS LINE was record delimiter
             else:
+                # print("(emptyline)")
                 if last_key in self._parameters:
                     if parameter["type"] == "delimiter":
                         if 'language_fullname' not in hyperdata.keys():
@@ -97,6 +114,7 @@ class RISParser(Parser):
                                 if 'language_iso2' not in hyperdata.keys():
                                     hyperdata['language_iso2'] = 'en'
                         yield hyperdata
+                        # print("{saved previous record}")
                         last_key = None
                         hyperdata = {}
             # [end of loop per lines]
@@ -111,6 +129,7 @@ class RISParser(Parser):
                     hyperdata[parameter["key"]] = final_value
                 else:
                     hyperdata = PY_values_decompose_and_save(final_value, hyperdata)
+                # print("{saved previous"+last_key+"}")
 
         # if a hyperdata object is left in memory, yield it as well
         if hyperdata:
@@ -119,8 +138,7 @@ class RISParser(Parser):
                     if 'language_iso2' not in hyperdata.keys():
                         hyperdata['language_iso2'] = 'en'
             yield hyperdata
-
-
+            # print("{saved previous record}")