From 54596f455016d22fa34b4e7dba8d3cfea053f3d9 Mon Sep 17 00:00:00 2001
From: Pieter De Schepper <pieter@netlog.com>
Date: Thu, 16 Jan 2020 12:53:54 +0100
Subject: [PATCH 1/3] make sure a custom seperator can also be used instead of
 wrapping in <span> tags

---
 budou/budou.py | 14 +++++++++++---
 budou/chunk.py |  6 ++++++
 2 files changed, 17 insertions(+), 3 deletions(-)

diff --git a/budou/budou.py b/budou/budou.py
index f90caaa..1ea14a5 100644
--- a/budou/budou.py
+++ b/budou/budou.py
@@ -17,7 +17,7 @@
 """Budou: an automatic organizer tool for beautiful line breaking in CJK
 
 Usage:
-  budou [--segmenter=<seg>] [--language=<lang>] [--classname=<class>] [--inlinestyle] [--wbr] [<source>]
+  budou [--segmenter=<seg>] [--language=<lang>] [--seperator=<seperator>] [--classname=<class>] [--inlinestyle] [--wbr] [<source>]
   budou -h | --help
   budou -v | --version
 
@@ -31,6 +31,9 @@
 
   --language=<language>       Language the source in.
 
+  --seperator=<seperator>     Custom seperator instead of SPAN tags, when used
+                              classname and inlinestyle are ignored
+
   --classname=<classname>     Class name for output SPAN tags. Use
                               comma-separated value to specify multiple classes.
 
@@ -75,10 +78,15 @@ def main():
       inlinestyle=args['--inlinestyle'],
       wbr=args['--wbr'],
       )
-  print(result['html_code'].encode('utf-8'))
+
+  if args['--seperator']:
+    print(result['chunks'].seperator_serialize(args['--seperator']).encode('utf-8'))
+  else:
+    print(result['html_code'].encode('utf-8'))
+
   sys.exit()
 
-def parse(source, segmenter='nlapi', language=None, max_length=None,
+def parse(source, segmenter='nlapi', language=None, max_length=None, seperator=None,
           classname=None, attributes=None, inlinestyle=False, wbr=False,
           **kwargs):
   """Parses input source.
diff --git a/budou/chunk.py b/budou/chunk.py
index cf1f909..215ba1d 100644
--- a/budou/chunk.py
+++ b/budou/chunk.py
@@ -373,3 +373,9 @@ def wbr_serialize(self):
         ))
     return result
 
+  def seperator_serialize(self, seperator):
+    result =[]
+    for chunk in self:
+      result.append(chunk.word)
+    return seperator.join(result)
+

From 18da49c262fb8c7a8d69fc9d4fd98a2a8c6dbb3a Mon Sep 17 00:00:00 2001
From: Pieter De Schepper <pieter.schepper@gmail.com>
Date: Thu, 16 Jan 2020 12:53:54 +0100
Subject: [PATCH 2/3] make sure a custom seperator can also be used instead of
 wrapping in <span> tags

---
 budou/budou.py | 14 +++++++++++---
 budou/chunk.py |  6 ++++++
 2 files changed, 17 insertions(+), 3 deletions(-)

diff --git a/budou/budou.py b/budou/budou.py
index f90caaa..1ea14a5 100644
--- a/budou/budou.py
+++ b/budou/budou.py
@@ -17,7 +17,7 @@
 """Budou: an automatic organizer tool for beautiful line breaking in CJK
 
 Usage:
-  budou [--segmenter=<seg>] [--language=<lang>] [--classname=<class>] [--inlinestyle] [--wbr] [<source>]
+  budou [--segmenter=<seg>] [--language=<lang>] [--seperator=<seperator>] [--classname=<class>] [--inlinestyle] [--wbr] [<source>]
   budou -h | --help
   budou -v | --version
 
@@ -31,6 +31,9 @@
 
   --language=<language>       Language the source in.
 
+  --seperator=<seperator>     Custom seperator instead of SPAN tags, when used
+                              classname and inlinestyle are ignored
+
   --classname=<classname>     Class name for output SPAN tags. Use
                               comma-separated value to specify multiple classes.
 
@@ -75,10 +78,15 @@ def main():
       inlinestyle=args['--inlinestyle'],
       wbr=args['--wbr'],
       )
-  print(result['html_code'].encode('utf-8'))
+
+  if args['--seperator']:
+    print(result['chunks'].seperator_serialize(args['--seperator']).encode('utf-8'))
+  else:
+    print(result['html_code'].encode('utf-8'))
+
   sys.exit()
 
-def parse(source, segmenter='nlapi', language=None, max_length=None,
+def parse(source, segmenter='nlapi', language=None, max_length=None, seperator=None,
           classname=None, attributes=None, inlinestyle=False, wbr=False,
           **kwargs):
   """Parses input source.
diff --git a/budou/chunk.py b/budou/chunk.py
index cf1f909..215ba1d 100644
--- a/budou/chunk.py
+++ b/budou/chunk.py
@@ -373,3 +373,9 @@ def wbr_serialize(self):
         ))
     return result
 
+  def seperator_serialize(self, seperator):
+    result =[]
+    for chunk in self:
+      result.append(chunk.word)
+    return seperator.join(result)
+

From 1f46714f58e7a81cca67b646de96e0b514521ae9 Mon Sep 17 00:00:00 2001
From: Pieter De Schepper <pieter.schepper@gmail.com>
Date: Mon, 20 Jan 2020 14:20:42 +0100
Subject: [PATCH 3/3] Update budou.py

remove obsolete separator argument
---
 budou/budou.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/budou/budou.py b/budou/budou.py
index 03738a7..3780e7a 100644
--- a/budou/budou.py
+++ b/budou/budou.py
@@ -90,7 +90,7 @@ def main():
 
   sys.exit()
 
-def parse(source, segmenter='nlapi', language=None, max_length=None, separator=None,
+def parse(source, segmenter='nlapi', language=None, max_length=None,
           classname=None, attributes=None, inlinestyle=False, wbr=False,
           **kwargs):
   """Parses input source.