Adding function split_parts() to module pp_admintools.common

diff --git a/lib/pp_admintools/app/__init__.py b/lib/pp_admintools/app/__init__.py
index aa899dedecfdf2a25494a36a7d54873e1607cccc..2009e6c6ce314c68fef303bf663c61935246e00b 100644 (file)
--- a/lib/pp_admintools/app/__init__.py
+++ b/lib/pp_admintools/app/__init__.py
@@ -264,4 +264,4 @@ if __name__ == '__main__':
     pass
 
 
-# vim: ts=4 et list
+# vim: tabstop=4 expandtab shiftwidth=4 softtabstop=4 list

diff --git a/lib/pp_admintools/common.py b/lib/pp_admintools/common.py
index 5cc84c36e9474faf1e8d3da25d87ca685334140d..6ca850cb48bbcc3c808a199d4fe45a962ea6d745 100644 (file)
--- a/lib/pp_admintools/common.py
+++ b/lib/pp_admintools/common.py
@@ -19,6 +19,7 @@ from fb_tools.common import timeinterval2delta
 
 # Own modules
 from . import DEFAULT_TZ_OFFSET
+from .errors import BrokenStringSplit
 from .errors import DateFormatError
 from .errors import WrongDateIsoformatError
 from .xlate import XLATOR
@@ -33,7 +34,7 @@ except ImportError:
 _ = XLATOR.gettext
 ngettext = XLATOR.ngettext
 
-__version__ = '0.1.0'
+__version__ = '0.2.0'
 
 LOG = logging.getLogger(__name__)
 
@@ -51,6 +52,16 @@ RE_DEFAULT_TZ = re.compile(PATTERN_DEFAULT_TZ)
 
 RE_TZ = re.compile(r'^\s*(?P<tz_hours>[01]\d)(?::?(?P<tz_mins>[0-5]\d))?')
 
+RE_SPLIT_WS_BEGIN = re.compile(r'^\s+')
+RE_SPLIT_SQ = re.compile(r"\\'")
+RE_SPLIT_SQ_CHUNK = re.compile(r"^'((?:\\'|[^'])*)'")
+RE_SPLIT_DQ = re.compile(r'\\"')
+RE_SPLIT_DQ_CHUNK = re.compile(r'^"((?:\\"|[^"])*)"')
+RE_SPLIT_UQ_CHUNK = re.compile(r'^((?:[^\s\'"]+|\\\'|\\")+)')
+RE_SPLIT_WS_ALL = re.compile(r'^\s*$')
+RE_SPLIT_UNBALANCED = re.compile(r'^([\'"].*)\s*')
+
+
 DEFAULT_TZ = UTC
 
 # =============================================================================
@@ -143,6 +154,96 @@ def fromisoformat(datestr):
     return datetime.datetime(**params)
 
 
+# ==============================================================================
+def split_parts(text, keep_quotes=False, raise_on_unbalanced=True):
+    """
+    Split the given text in chunks by whitespaces or single or double quoted strings.
+
+    @param text: the text to split in chunks
+    @type text: str
+    @param keep_quotes: keep quotes of quoted chunks
+    @type keep_quotes: bool
+    @param raise_on_unbalanced: raise an exception on unbalanced quotes
+    @type raise_on_unbalanced: bool
+
+    @return: list of chunks
+    @rtype: list
+    """
+    chunks = []
+    if text is None:
+        return chunks
+
+    txt = str(text)
+    last_chunk = ''
+
+    # Big loop to split the text - until it is empty
+    while txt != '':
+
+        # add chunk, if there is a chunk left and a whitspace at the begin of the line
+        match = RE_SPLIT_WS_BEGIN.search(txt)
+        if (last_chunk != '') and match:
+            chunks.append(last_chunk)
+            last_chunk = ''
+
+        # clean the line
+        txt = txt.strip()
+        if txt == '':
+            break
+
+        # search for a single quoted string at the begin of the line
+        match = RE_SPLIT_SQ_CHUNK.search(txt)
+        if match:
+            chunk = match.group(1)
+            chunk = RE_SPLIT_SQ.sub("'", chunk)
+            if keep_quotes:
+                chunk = "'" + chunk + "'"
+            last_chunk += chunk
+            txt = RE_SPLIT_SQ_CHUNK.sub('', txt)
+            continue
+
+        # search for a double quoted string at the begin of the line
+        match = RE_SPLIT_DQ_CHUNK.search(txt)
+        if match:
+            chunk = match.group(1)
+            chunk = RE_SPLIT_DQ('"', chunk)
+            if keep_quotes:
+                chunk = '"' + chunk + '"'
+            last_chunk += chunk
+            txt = RE_SPLIT_DQ_CHUNK.sub('', txt)
+            continue
+
+        # search for unquoted, whitespace delimited text at the begin of the line
+        match = RE_SPLIT_UQ_CHUNK.search(txt)
+        if match:
+            last_chunk += match.group(1)
+            txt = RE_SPLIT_UQ_CHUNK.sub('', txt)
+            continue
+
+        # Only whitespaces left
+        if RE_SPLIT_WS_ALL.search(txt):
+            break
+
+        # Check for unbalanced quotes
+        match = RE_SPLIT_UNBALANCED.search(txt)
+        if match:
+            chunk = match.group(1)
+            if raise_on_unbalanced:
+                msg = _('Unbalanced quotes in {!r}.').format(text)
+                raise BrokenStringSplit(msg)
+            else:
+                last_chunk += chunk
+                continue
+
+        # Here we should not come to ...
+        msg = _('Broken split of {text!r}: {txt!r} left.').format(text=text, txt=txt)
+        raise BrokenStringSplit(msg)
+
+    if last_chunk != '':
+        chunks.append(last_chunk)
+
+    return chunks
+
+
 # =============================================================================
 
 if __name__ == '__main__':

diff --git a/lib/pp_admintools/errors.py b/lib/pp_admintools/errors.py
index 2eec508903548980c8878e42bcaf395281b92d6b..24adf235654a0f7343993eaca4c8ac1f7c652550 100644 (file)
--- a/lib/pp_admintools/errors.py
+++ b/lib/pp_admintools/errors.py
@@ -14,7 +14,7 @@ from .xlate import XLATOR
 
 _ = XLATOR.gettext
 
-__version__ = '0.9.1'
+__version__ = '0.10.0'
 
 
 # =============================================================================
@@ -24,6 +24,13 @@ class DpxError(FbError):
     pass
 
 
+# =============================================================================
+class BrokenStringSplit(DpxError):
+    """Error class on an exception inside common.split_parts()."""
+
+    pass
+
+
 # =============================================================================
 class DpxAppError(DpxError, FbAppError):
     """Base error class for all self defined exceptions in applications."""