support subsections (fix #10)

src/doc2tosca.py

@@ -33,9 +33,21 @@ metadata:
 imports:
   {imports}
 
-data_types:
 '''
 
+SUBSECTIONS = [
+    "Artifact Types",
+    "Data Types",
+    "Capability Types",
+    "Interface Types",
+    "Requirements Types",
+    "Relationship Types",
+    "Interface Types",
+    "Node Types",
+    "Group Types",
+    "Policy Types"
+]
+
 MODELS = {}
 EXAMPLES = {}
 
@@ -109,8 +121,8 @@ def get_content(doc):
         elif isinstance(element, docx.oxml.table.CT_Tbl):
             ret.append(Table(element, body))
             table_count = table_count + 1
-        else:
-            print("Non paragraph or table " +  str(type(element)))
+        #else:
+        #    print("Non paragraph or table " +  str(type(element)))
     print("Paragraphs: " + str(parag_count))
     print("Tables: " + str(table_count))
     return ret
@@ -130,6 +142,17 @@ def find_sect(sect_to_find, start_idx, doc_content):
     print("FOUND " + sect_to_find + " at " + str(start_idx))
     return start_idx
 
+def is_lvl2_section_hdn(txt):
+    ''' Returns true if txt is level 2 heading'''
+    clean_txt = txt.strip()
+    
+    if not bool(re.match(r'^[0-9]\.[0-9]+\t[a-zA-Z\s]*$', clean_txt)):
+        return False
+
+    subtitle = clean_txt.split('\t')[1]
+    
+    return subtitle in SUBSECTIONS
+
 def is_lvl1_section_hdn(txt):
     ''' Returns true if txt is level 1 heading'''
     clean_txt = txt.strip()
@@ -159,6 +182,15 @@ def find_all_sections(doc_content):
     sections.append(Section(start_indx, end_indx-1, doc_content[start_indx].text))
     return sections
 
+def write_subsection_to_file(txt, buf):
+    '''
+    Writes a subsection header in utf-8 encoding to file buf
+    '''
+    buf.write(slugify(txt)+":")
+    if not txt.endswith('\n'):
+        buf.write('\n')
+    buf.write('\n')
+
 def write_table_to_file(tab, buf):
     '''
     Writes content of table t in utf-8 encoding to file F
@@ -177,28 +209,28 @@ def gen_tables_btwn(a_id, b_id, content, buf):
     definitions_count = 0
 
     for idx in range(a_id, b_id):
+
         if idx >= len(content):
-            print("A: " + str(a_id))
-            print("B: " + str(b_id))
-            print("IDX: " + str(idx))
-            print("LEN(CONTENT): " + str(len(content)))
-            return definitions_count 
+            print("ERROR: Paragraph out of range (A: {}, B: {}, IDX: {}, LEN(CONTENT): {})".format(
+                a_id, b_id,idx, len(content)))
+            return definitions_count
+
         tmp_elem = content[idx]
+
+        if isinstance(tmp_elem, Paragraph) and is_lvl2_section_hdn(tmp_elem.text):
+            print(tmp_elem.text)
+            write_subsection_to_file(tmp_elem.text.split("\t")[1], buf)
+            definitions_count = definitions_count + 1
+
         if isinstance(tmp_elem, Table) and is_tosca_def(tmp_elem):
             write_table_to_file(tmp_elem, buf)
             definitions_count = definitions_count + 1
+
         elif isinstance(tmp_elem, Table):
             txt = tmp_elem.rows[0].cells[0].text
             if txt.strip().startswith("Name") or txt.strip().startswith("Shorthand") or \
                 txt.strip().startswith("tosca_def"):
                 continue
-            # print("----- Filtered out: " + txt.split("\n")[0])
-            #if not len(tmp_elem.rows) == 1:
-                #print("       Rows count != 1 ")
-            #if not len(tmp_elem.columns) == 1:
-            #    print("       Columns count != 1 ")
-            #if not match_definition_incipit(txt):
-            #    print("       Regex != 1 ")
     return definitions_count
 
 def generate_header(
@@ -277,7 +309,7 @@ def generate_templates(
 
 
 def print_to_files(prefix=None):
-    '''
+    ''' 
     Prefix is a path to a folder to work into
     '''
     for key in MODELS:
@@ -317,6 +349,9 @@ def parse_version_from_filename(filename):
                 .replace("0",".").strip(".").strip("p.docx")
     return ""
 
+def slugify(t):
+    return t.replace(" ", "_").lower()
+
 if __name__ == "__main__":
 
     try:

src/test_doc2tosca.py

@@ -20,6 +20,12 @@ def test_is_lvl1_section_hdn():
     assert d2t.is_lvl1_section_hdn("Annex A (informative)")
     assert d2t.is_lvl1_section_hdn("Annex C (normative):\tConformance\t284")
 
+def test_is_lvl2_section_hdn():
+    
+    assert d2t.is_lvl2_section_hdn("6.3\tData Types")
+    assert not d2t.is_lvl2_section_hdn("6.4.2\tSomething")
+    assert not d2t.is_lvl2_section_hdn("6.4\tSomething")
+
 def test_section_init():
 
     ssss = d2t.Section(0, 10, "6\tVNFD TOSCA model")