From c7a089418229cee6d2cc6b6f8f1e32e89af42aaa Mon Sep 17 00:00:00 2001
From: "yusuke.shinyama.dummy"
 <yusuke.shinyama.dummy@1aa58f4a-7d42-0410-adbc-911cccaed67c>
Date: Sat, 20 Jun 2009 10:00:51 +0000
Subject: [PATCH] auto detect output type

git-svn-id: https://pdfminerr.googlecode.com/svn/trunk/pdfminer@115 1aa58f4a-7d42-0410-adbc-911cccaed67c
---
 Makefile         |  3 ++-
 tools/pdf2txt.py | 19 ++++++++++++++++---
 2 files changed, 18 insertions(+), 4 deletions(-)

diff --git a/Makefile b/Makefile
index 6010a59..6630b23 100644
--- a/Makefile
+++ b/Makefile
@@ -5,6 +5,7 @@ PACKAGE=pdfminer
 SVN=svn
 GNUTAR=tar
 PYTHON=python
+PREFIX=/usr/local
 TMPDIR=/tmp
 VERSION=`$(PYTHON) $(PACKAGE)/__init__.py`
 DISTNAME=$(PACKAGE)-dist-$(VERSION)
@@ -15,7 +16,7 @@ CONV_CMAP=$(PYTHON) pdfminer/cmap.py
 all:
 
 install:
-	$(PYTHON) setup.py install
+	$(PYTHON) setup.py install --prefix=$(PREFIX)
 
 clean:
 	-rm -rf build
diff --git a/tools/pdf2txt.py b/tools/pdf2txt.py
index 39d283d..8a5086e 100755
--- a/tools/pdf2txt.py
+++ b/tools/pdf2txt.py
@@ -26,9 +26,9 @@ def main(argv):
   pagenos = set()
   maxpages = 0
   # output option
-  outtype = 'html'
+  outfile = None
+  outtype = None
   codec = 'utf-8'
-  outfp = sys.stdout
   cluster_margin = None
   pageno = 1
   scale = 1
@@ -41,7 +41,7 @@ def main(argv):
     elif k == '-m': maxpages = int(v)
     elif k == '-t': outtype = v
     elif k == '-c': codec = v
-    elif k == '-o': outfp = file(v, 'wb')
+    elif k == '-o': outfile = v
     elif k == '-s': scale = float(v)
     elif k == '-T': cluster_margin = float(v)
   #
@@ -54,6 +54,19 @@ def main(argv):
   #
   CMapDB.initialize(cmapdir)
   rsrc = PDFResourceManager()
+  if not outtype:
+    outtype = 'text'
+    if outfile:
+      if outfile.endswith('.htm') or outfile.endswith('.html'):
+        outtype = 'html'
+      elif outfile.endswith('.sgml'):
+        outtype = 'sgml'
+      elif outfile.endswith('.tag'):
+        outtype = 'tag'
+  if outfile:
+    outfp = file(outfile, 'w')
+  else:
+    outfp = sys.stdout
   if outtype == 'sgml':
     device = SGMLConverter(rsrc, outfp, codec=codec, cluster_margin=cluster_margin)
   elif outtype == 'html':