Support for processing a filename rather than a section, and add --field-separator

.gitignore

@@ -0,0 +1,12 @@
+/pgloader/__init__.pyc
+/pgloader/csvreader.pyc
+/pgloader/db.pyc
+/pgloader/fixedreader.pyc
+/pgloader/lo.pyc
+/pgloader/logger.pyc
+/pgloader/options.pyc
+/pgloader/pgloader.pyc
+/pgloader/reader.pyc
+/pgloader/textreader.pyc
+/pgloader/tools.pyc
+/pgloader.1.xml

debian/changelog

@@ -1,3 +1,10 @@
+pgloader (2.3.3-1) unstable; urgency=low
+
+  * Implement -f --field-sep to overwrite the default from command line
+  * Add support for filename arguments, which use defaults
+
+ -- Dimitri Fontaine <dim@tapoueh.org>  Sun,  4 Apr 2010 19:34:39 +0200
+
 pgloader (2.3.2-1) unstable; urgency=low
 
   * Implement --from support in all readers (Closes: #531034)

pgloader.1.txt

@@ -11,7 +11,7 @@ pgloader [--version] [-c configuration file]
 	 [-l loglevel] [-L logfile]
          [-n dryrun] [-Cn count] [-Fn from] [-In from id]
          [-E input files encoding] [-R reformat:path]
-         [Section1 Section2]
+         [Section|Filename ...] 
 
 == DESCRIPTION ==
 
@@ -42,6 +42,28 @@ errors is issued.
 http://pgfoundry.org/projects/pgloader/[], where you'll find a debian
 package, a source package and an anonymous CVS.
 
+== Arguments ==
+
++pgloader+ as of version +2.3.3+ accepts two kinds of arguments, either
+section names of file names. If both a section and a file exist with the
+same name, preference is given to the section, where you can edit your
+settings rather than using default ones.
+
+Section::
++
+is the name of a configured Section describing some data to load
++
+Section arguments are optional, if no section is given all configured
+sections are processed.
+
+Filename::
+
+    The name of a file containing the data to load. +pgloader+ will
+    internally setup a +Section+ for this filename, with the default field
+    separator or the given +--field-separator+ and the +columns+ parameter
+    set to '*', and more importantly the format set to +CSV+. It's the only
+    supported format with sane enough defaults to apply here.
+
 == OPTIONS ==
 
 In order for pgloader to run, you have to edit a configuration file
@@ -135,6 +157,11 @@ preferring to use --from instead.
 +
 You can't use both -F and -I at the same time.
 
+-f, --field-sep::
+
+    Default field separator to use, when not set +pgloader+ will use
+    '|'. Useful when using +filename+ arguments rather than +section+ ones.
+
 -E, --encoding::
 
     Input data files encoding. Defaults to 'latin9'.
@@ -158,13 +185,6 @@ You can't use both -F and -I at the same time.
     Force +pgloader+ to use given version of psycopg, either +1+ or
     +2+.
 
-Section::
-+
-is the name of a configured Section describing some data to load
-+
-Section arguments are optional, if no section is given all configured
-sections are processed.
-
 == GLOBAL CONFIGURATION SECTION ==
 
 The configuration file has a +.ini+ file syntax, its first section has

pgloader.py

@@ -95,6 +95,10 @@ def parse_options():
                       default = None,
                       help    = "wait for given id on input to begin")
 
+    parser.add_option("-f", "--field-separator", dest = "fsep",
+                      default = pgloader.options.FIELD_SEP,
+                      help    = "default field separator to use")
+
     parser.add_option("-E", "--encoding", dest = "encoding",
                       default = None,
                       help    = "input files encoding")
@@ -183,6 +187,7 @@ def parse_options():
     pgloader.options.COUNT      = opts.count
     pgloader.options.FROM_COUNT = opts.fromcount
     pgloader.options.FROM_ID    = opts.fromid
+    pgloader.options.FIELD_SEP  = opts.fsep
 
     pgloader.options.INPUT_ENCODING = opts.encoding
 
@@ -482,13 +487,30 @@ def load_data():
     sections = []
     summary  = {}
 
-    # args are meant to be configuration sections
+    # args are meant to be configuration sections, or filenames
     if len(args) > 0:
         for s in args:
             if config.has_section(s):
                 sections.append(s)
+            else:
+                log.info("Creating a section for file '%s'" % s)
+                # a filename was given, apply [pgsql] defaults
+                # set the tablename as the filename sans extension
+                # consider columns = *
+                if not os.path.exists(s):
+                    print >>sys.stderr, \
+                        "Error: '%s' does not exists as a section nor as a file" % s
+                    sys.exit(2)
+
+                config.add_section(s)
+                config.set(s, 'table', os.path.splitext(os.path.basename(s))[0])
+                config.set(s, 'filename', s)
+                config.set(s, 'columns', '*')
+                config.set(s, 'format', 'csv')
+                sections.append(s)
 
     else:
+        log.debug("No argument on CLI, will consider all sections")
         for s in config.sections():
             if s != 'pgsql':
                 sections.append(s)