Merge pull request #332 from socialcopsdev/fix-312

[MRG] Fix #312

HISTORY.md

@@ -8,6 +8,7 @@ master
 
 * [#293](https://github.com/socialcopsdev/camelot/issues/293) Split text ignores all text to the right of last cut. [#294](https://github.com/socialcopsdev/camelot/pull/294) by Vinayak Mehta.
 * [#277](https://github.com/socialcopsdev/camelot/issues/277) Sort TableList by order of tables in PDF. [#283](https://github.com/socialcopsdev/camelot/pull/283) by [Sym Roe](https://github.com/symroe).
+* [#312](https://github.com/socialcopsdev/camelot/issues/312) `table_regions` throws `ValueError` when `flavor='stream'`. [#332](https://github.com/socialcopsdev/camelot/pull/332) by Vinayak Mehta.
 
 0.7.2 (2019-01-10)
 ------------------

camelot/parsers/stream.py

@@ -286,7 +286,11 @@ class Stream(BaseParser):
                 # filter horizontal text
                 hor_text = []
                 for region in self.table_regions:
-                    x1, y1, x2, y2 = region
+                    x1, y1, x2, y2 = region.split(",")
+                    x1 = float(x1)
+                    y1 = float(y1)
+                    x2 = float(x2)
+                    y2 = float(y2)
                     region_text = text_in_bbox((x1, y2, x2, y1), self.horizontal_text)
                     hor_text.extend(region_text)
             # find tables based on nurminen's detection algorithm

tests/test_common.py

@@ -69,6 +69,14 @@ def test_stream_two_tables():
     assert df2.equals(tables[1].df)
 
 
+def test_stream_table_regions():
+    df = pd.DataFrame(data_stream_table_areas)
+
+    filename = os.path.join(testdir, "tabula/us-007.pdf")
+    tables = camelot.read_pdf(filename, flavor="stream", table_regions=["320,460,573,335"])
+    assert df.equals(tables[0].df)
+
+
 def test_stream_table_areas():
     df = pd.DataFrame(data_stream_table_areas)