Add a massimport page

[djsite.git] / quotes / massimport.py

diff --git a/quotes/massimport.py b/quotes/massimport.py
new file mode 100644 (file)
index 0000000..ade6f38
--- /dev/null
+++ b/quotes/massimport.py
@@ -0,0 +1,95 @@
+import re
+from django.db import DatabaseError, transaction
+
+from quotes.models import QuoteTag, Quote, Work, Author
+
+def get_or_create_author(authorname, resultcontext):
+    authorname = authorname.strip()
+    try:
+        return Author.objects.get(name=authorname)
+    except Author.DoesNotExist:
+        a = Author.objects.create(name=authorname,
+                                  pvt_notes="<p>Created during mass import</p>")
+        resultcontext["created_authors"] += [a]
+        return a
+
+def get_or_create_work(workname, author, resultcontext):
+    workname = workname.strip()
+    try:
+        return Work.objects.get(name=workname, author=author)
+    except Work.DoesNotExist:
+        w = Work.objects.create(name=workname,
+                                author=author,
+                                pvt_notes="<p>Created during mass import</p>")
+        resultcontext["created_works"] += [w]
+        return w
+
+def get_or_create_tag(tagname, resultcontext):
+    try:
+        return QuoteTag.objects.get(tag=tagname)
+    except QuoteTag.DoesNotExist:
+        t = QuoteTag.objects.create(tag=tagname)
+        resultcontext["created_tags"] += [t]
+        return t
+    
+def add_tags_on_quote(quote, tagline, resultcontext):
+    for tagname in tagline.split(","):
+        tagname = tagname.strip()
+        if not tagname:
+            continue
+        tag = get_or_create_tag(tagname, resultcontext)
+        quote.tags.add(tag)
+
+def paragraphize(text):
+    paragraph = ""
+    for line in text.splitlines():
+        line = line.strip()
+        if not line: continue
+        paragraph += "<p>%s</p>" % line
+        # rest of the HTML will be bleach.clean()'d
+    return paragraph
+        
+def create_quote(quotetext, authorname, workname, tagline, resultcontext):
+    author = get_or_create_author(authorname, resultcontext)
+    work = get_or_create_work(workname, author, resultcontext)
+    quotetext = paragraphize(quotetext)
+    quote = Quote.objects.create(text=quotetext, work=work)
+    add_tags_on_quote(quote, tagline, resultcontext)
+    resultcontext["created_quotes"] += [quote]
+
+def create_all_quotes(allquotes, resultcontext):
+    quotesep = re.compile(r'^\s*===+.*$', re.MULTILINE)
+    elemsep = re.compile(r'^\s*---+.*$', re.MULTILINE)
+    
+    for fullquote in quotesep.split(allquotes):
+        if not fullquote: continue
+
+        elements = elemsep.split(fullquote)
+        if len(elements) not in [3, 4]:
+            resultcontext["rejected"] += [fullquote]
+            continue
+        quotetext, authorname, workname = elements[0:3]
+        if len(elements) == 4:
+            tagline = elements[3]
+        else:
+            tagline = ''
+        create_quote(quotetext, authorname, workname, tagline,
+                     resultcontext)
+
+def domassimport(allquotes):
+    resultcontext = {}
+    resultcontext["created_quotes"] = []
+    resultcontext["created_tags"] = []
+    resultcontext["created_works"] = []
+    resultcontext["created_authors"] = []
+    resultcontext["rejected"] = []
+    resultcontext["fatal_error"] = False
+
+    try:
+        with transaction.atomic():
+            create_all_quotes(allquotes, resultcontext)
+    except DatabaseError as e:
+        resultcontext["fatal_error"] = True
+        resultcontext["fatal_error_message"] = e.__cause__
+
+    return resultcontext