starting a modified environment for branch 1.1

README.org

@@ -52,6 +52,10 @@ Pipelet is a free framework which helps you :
   adapt to different hardware and software architectures.
 
 
+*** What's new in v1.1
+- The glob_seg behavior has been modified for coherence, convenience,
+  and performance sake. See [[*The%20segment%20environment][The segment environment]].
+
 ** Getting started
 *** Pipelet installation 
 **** Dependencies 
@@ -121,7 +125,6 @@ This command ends up with the creation of directory named pipename wich contains
 
 The next section describes those two files in more details. 
 
-
 ** Writing Pipes
 *** Pipeline architecture
 
@@ -348,8 +351,16 @@ The segment code is executed in a specific environment that provides:
 
 2. Functionalities to use the automated hierarchical data storage system.
    - =get_data_fn(basename)=: complete the filename with the path to the working directory. 
-   - =glob_seg(seg, regexp)=: return the list of filename in segment seg
-     working directory matching regexp.
+   - =glob_seg(seg, regexp)=: Return the list of filename matching the pattern y in the
+        data directory of parent tasks from the parent segment x.
+   - =glob_seg_all(seg, regexp)=: Return the list of filename matching
+        y in the working directory of segment x independantly of
+        whether the file comes from a task related to the current
+        task. glob_seg_all is provided to reproduce the behaviour of
+        old glob_seg for backward compatibility. Its usage should be limited as it:
+        - potentially breaks the dependancy scheme.
+        - may hurt performances as all task directories of the segment
+          x will be searched.
    - =get_tmp_fn()=: return a temporary filename.
 
 3. Functionalities to use the automated parameters handling
@@ -1169,7 +1180,22 @@ And the corresponding cgi script:
 #+end_src
 
 
-* The Pipelet actors
+* Contributing to pipelet
+** Improvements suggestion
+As far as possible, we try to keep pipelet ReadMe driven. If you want
+to ask for new features or modifications to pipelet, a good way to do
+it is under the form of a patch to the README.org file. If you are
+using git:
+- Figure out how actually would like pipelet to behave
+- Edit the README.org in consequence. Org files are simple text files
+  that can be conveniently edited using emacs org-mode.
+- Commit your proposition with a relevant commit message.
+  =git commit -a=
+- Format the patch for email submission, and send it to us
+  =git send-email --to  HEAD^=
+
+
+** The Pipelet actors
   
 This section document the code for developers.  The code documentation
 can be built using the doxygen configuration file

pipelet/environment.py

@@ -182,12 +182,33 @@ class Environment(EnvironmentBase):
         self.logger.info ("hooking %s"%hook_name)
         return self._hook(hook_name, glo)
     
-    
-    
     def glob_seg(self, x, y):
+        """ Globbing limited to the fatherhood
+
+        For unlimited globbing see glob_seg_all.
+        
+        Parameters
+        ----------
+        x: string, segment name
+        y: string, globbing pattern
+        
+        Returns
+        -------
+        Return the list of filename matching the pattern y in the
+        data directory of parent tasks from the parent segment x.
+        """
+        segx = self._worker.pipe.find_seg(self._worker.task.seg, x)
+        if segx is None:
+            self.logger.warning("No parent segment matching %s found"%x)
+        
+    def glob_seg_all(self, x, y):
         """ Return the list of filename matching y in the working
         directory of segment x.
         
+        Usage of glob_seg_all should be limited:
+        - potentially breaks the dependancy scheme
+        - May hurt performances as all task directories of the segment x will be searched
+
         Parameters
         ----------
         x: string, segment name
@@ -205,7 +226,7 @@ class Environment(EnvironmentBase):
             +glob(path.join(self._worker.pipe.get_data_dir(segx),path.join('*/',y)))
 
     
-    def logged_subprocess(self, args, shell=False, except_on_failure=True):
+    def logged_subprocess(self, args, shell=False, except_on_failure=True, name=None):
         """ Execute a subprocess and log its output.
         
         Create files process_name.log and process_name.err
@@ -221,8 +242,10 @@ class Environment(EnvironmentBase):
         tuple (outputfile, errorfile)
         """
         proc = args[0]
-        of = self.get_data_fn(proc+'.log')
-        ef = self.get_data_fn(proc+'.err')
+        if name is None:
+            name = proc
+        of = self.get_data_fn(name+'.log')
+        ef = self.get_data_fn(name+'.err')
         o = file(of,'w')
         e = file(ef,'w')
         e.write('#'+' '.join([str(a) for a in args])+'\n')