From 12f36e938d32c3648e7636cea861d0a59333d007 Mon Sep 17 00:00:00 2001
From: Edward Betts <edward@4angle.com>
Date: Mon, 20 Aug 2018 13:33:18 +0100
Subject: [PATCH] Use first line of sourcedoc as title

---
 sourcing/model.py  | 23 +++++++++++++++++------
 sourcing/text.py   |  6 ++++++
 tests/test_text.py | 21 +++++++++++++++++++++
 3 files changed, 44 insertions(+), 6 deletions(-)
 create mode 100644 tests/test_text.py

diff --git a/sourcing/model.py b/sourcing/model.py
index f5366bf..5e9286d 100644
--- a/sourcing/model.py
+++ b/sourcing/model.py
@@ -1,6 +1,7 @@
 from flask import url_for
 from .database import session
 from .parse import parse_link, parse_sourcedoc_facet, parse_span
+from .text import first_non_empty_line
 from sqlalchemy.ext.declarative import declarative_base
 from sqlalchemy import Column, ForeignKey
 from sqlalchemy.types import String, Unicode, Integer, DateTime, Boolean, UnicodeText, Enum
@@ -203,18 +204,18 @@ class Item(TimeStampedModel):
     def set_title_url(self):
         return self.view_url('set_title')
 
-    def title(self, titles=None):
+    def title_from_link(self, titles=None):
         if not titles:
             titles = XanaLink.get_all_titles()
-        return self.type + ": " + titles.get(self, self.hashid)
+        return titles.get(self)
+
+    def title(self, titles=None):
+        return self.type + ': ' + (self.title_from_link(titles) or self.hashid)
 
     def has_title(self):
         titles = XanaLink.get_all_titles()
         return self in titles
 
-    def title_from_link(self):
-        return XanaLink.get_all_titles().get(self)
-
     def set_title(self, title, user):
         title_source_doc = SourceDoc(text=title, user=user)
         session.add(title_source_doc)
@@ -278,7 +279,7 @@ class XanaLink(Item):
         if titles is None:
             titles = XanaLink.get_all_titles()
         if self in titles:
-            return self.type + ": " + titles[self]
+            return self.type + ': ' + titles[self]
 
         parsed = self.parse()
 
@@ -348,5 +349,15 @@ class SourceDoc(Item):
         result = s[:length - len(end)].rsplit(' ', 1)[0]
         return result + end
 
+    def title(self, titles=None):
+        titles = XanaLink.get_all_titles()
+        from_link = self.title_from_link(titles=titles)
+        if from_link:
+            return self.type + ': ' + from_link
+        first_line = first_non_empty_line(self.text)
+        if first_line:
+            return self.type + ': ' + first_line
+        return self.type + ': ' + self.hashid
+
 
 configure_mappers()
diff --git a/sourcing/text.py b/sourcing/text.py
index b8cb383..51fdaa3 100644
--- a/sourcing/text.py
+++ b/sourcing/text.py
@@ -39,3 +39,9 @@ def add_highlight(text, span_start, span_length):
             cur.append({'text': after})
 
         yield start, cur
+
+def first_non_empty_line(text):
+    for start, cur in iter_lines(text):
+        tidy = cur.strip()
+        if tidy:
+            return tidy
diff --git a/tests/test_text.py b/tests/test_text.py
new file mode 100644
index 0000000..e81eb77
--- /dev/null
+++ b/tests/test_text.py
@@ -0,0 +1,21 @@
+from sourcing import text
+
+def test_find_newlines():
+    sample = ''
+    assert list(text.find_newlines(sample)) == []
+
+    sample = 'aaa\nbbb\nccc\n'
+    assert list(text.find_newlines(sample)) == [4, 8, 12]
+
+def test_iter_lines():
+    sample = ''
+    result = list(text.iter_lines(sample))
+    assert result == []
+
+    sample = 'aaa\nbbb\nccc\n'
+    result = list(text.iter_lines(sample))
+    assert result == [(0, 'aaa\n'), (4, 'bbb\n'), (8, 'ccc\n')]
+
+def test_first_non_empty_line():
+    assert text.first_non_empty_line('') is None
+    assert text.first_non_empty_line('  \n  \n') is None