From c2d606bc6f604c6fbd89c99cfaea9150cd41f652 Mon Sep 17 00:00:00 2001
From: Friedrich Lindenberg <friedrich@pudo.org>
Date: Thu, 4 Apr 2013 23:05:19 +0200
Subject: [PATCH] Import datafreeze source code.

---
 dataset/freeze/__init__.py        |  0
 dataset/freeze/app.py             | 37 ++++++++++++++
 dataset/freeze/config.py          | 82 +++++++++++++++++++++++++++++++
 dataset/freeze/engine.py          | 42 ++++++++++++++++
 dataset/freeze/format/__init__.py | 14 ++++++
 dataset/freeze/format/common.py   | 76 ++++++++++++++++++++++++++++
 dataset/freeze/format/fcsv.py     | 37 ++++++++++++++
 dataset/freeze/format/fjson.py    | 45 +++++++++++++++++
 dataset/freeze/format/ftabson.py  | 24 +++++++++
 dataset/util.py                   | 54 ++++++++++++++++++++
 setup.py                          | 11 +++--
 11 files changed, 419 insertions(+), 3 deletions(-)
 create mode 100644 dataset/freeze/__init__.py
 create mode 100644 dataset/freeze/app.py
 create mode 100644 dataset/freeze/config.py
 create mode 100644 dataset/freeze/engine.py
 create mode 100644 dataset/freeze/format/__init__.py
 create mode 100644 dataset/freeze/format/common.py
 create mode 100644 dataset/freeze/format/fcsv.py
 create mode 100644 dataset/freeze/format/fjson.py
 create mode 100644 dataset/freeze/format/ftabson.py
 create mode 100644 dataset/util.py

diff --git a/dataset/freeze/__init__.py b/dataset/freeze/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/dataset/freeze/app.py b/dataset/freeze/app.py
new file mode 100644
index 0000000..1f57700
--- /dev/null
+++ b/dataset/freeze/app.py
@@ -0,0 +1,37 @@
+import logging
+import argparse
+
+from dataset.util import FreezeException
+from dataset.freeze.config import Configuration
+from dataset.freeze.engine import ExportEngine
+from dataset.freeze.format import get_serializer
+
+
+logging.basicConfig(level=logging.DEBUG)
+log = logging.getLogger(__name__)
+
+parser = argparse.ArgumentParser(
+        description='Generate static JSON and CSV extracts from a SQL database.',
+        epilog='For further information, please check the documentation.')
+parser.add_argument('config', metavar='CONFIG', type=str,
+                   help='freeze file cofiguration')
+
+def main():
+    try: 
+        args = parser.parse_args()
+        config = Configuration(args.config)
+        for export in config.exports:
+            if export.skip:
+                log.info("Skipping: %s", export.name)
+                continue
+            log.info("Running: %s", export.name)
+            engine = ExportEngine(export)
+            query = engine.query()
+            serializer_cls = get_serializer(export)
+            serializer = serializer_cls(engine)
+            serializer.serialize()
+    except FreezeException, fe:
+        log.error(fe)
+
+if __name__ == '__main__':
+    main()
diff --git a/dataset/freeze/config.py b/dataset/freeze/config.py
new file mode 100644
index 0000000..bb5756c
--- /dev/null
+++ b/dataset/freeze/config.py
@@ -0,0 +1,82 @@
+import json
+import yaml
+
+
+TRUISH = ['true', 'yes', '1', 'on']
+
+DECODER = {
+    'json': json,
+    'yaml': yaml
+    }
+
+
+def merge_overlay(data, overlay):
+    out = overlay.copy()
+    for k, v in data.items():
+        if isinstance(v, dict) and isinstance(out.get(k), dict):
+            v = merge_overlay(v, out.get(k))
+        out[k] = v
+    return out
+
+
+class Configuration(object):
+
+    def __init__(self, file_name):
+        self.file_name = file_name
+        extension = file_name.rsplit('.', 1)[-1]
+        loader = DECODER.get(extension, json)
+        try:
+            fh = open(file_name, 'rb')
+            try:
+                self.data = loader.load(fh)
+            except ValueError, ve:
+                raise FreezeException("Invalid freeze file: %s" % ve)
+            fh.close()
+        except IOError, ioe:
+            raise FreezeException(unicode(ioe))
+
+    @property
+    def exports(self):
+        if not isinstance(self.data, dict):
+            raise FreezeException("The root element of the freeze file needs to be a hash")
+        if not isinstance(self.data.get('exports'), list):
+            raise FreezeException("The freeze file needs to have a list of exports")
+        common = self.data.get('common', {})
+        for export in self.data.get('exports'):
+            yield Export(common, export)
+
+
+class Export(object):
+
+    def __init__(self, common, data):
+        self.data = merge_overlay(data, common)
+
+    def get(self, name, default=None):
+        return self.data.get(name, default)
+
+    def get_normalized(self, name, default=None):
+        value = self.get(name, default=default)
+        if not value in [None, default]:
+            value = unicode(value).lower().strip()
+        return value
+
+    def get_bool(self, name, default=False):
+        value = self.get_normalized(name)
+        if value is None:
+            return default
+        return value in TRUISH
+
+    def get_int(self, name, default=None):
+        value = self.get_normalized(name)
+        if value is None:
+            return default
+        return int(value)
+
+    @property
+    def skip(self):
+        return self.get_bool('skip')
+    
+    @property
+    def name(self):
+        return self.get('name', self.get('query'))
+
diff --git a/dataset/freeze/engine.py b/dataset/freeze/engine.py
new file mode 100644
index 0000000..3fcf229
--- /dev/null
+++ b/dataset/freeze/engine.py
@@ -0,0 +1,42 @@
+from sqlalchemy import create_engine
+from sqlalchemy.exc import ProgrammingError
+
+from dataset.util import FreezeException
+
+class Query(object):
+
+    def __init__(self, query, rp):
+        self.query = query
+        self.rp = rp
+
+    def __len__(self):
+        return self.rp.rowcount
+
+    def __iter__(self):
+        keys = self.rp.keys()
+        while True:
+            row = self.rp.fetchone()
+            if row is None:
+                return
+            yield dict(zip(keys, row))
+
+
+class ExportEngine(object):
+
+    def __init__(self, config):
+        self.config = config
+
+    @property
+    def engine(self):
+        if not hasattr(self, '_engine'):
+            self._engine = create_engine(self.config.get('database'))
+        return self._engine
+
+    def query(self):
+        try:
+            q = self.config.get('query')
+            rp = self.engine.execute(q)
+            return Query(q, rp)
+        except ProgrammingError, pe:
+            raise FreezeException("Invalid query: %s - %s" % (q, pe))
+
diff --git a/dataset/freeze/format/__init__.py b/dataset/freeze/format/__init__.py
new file mode 100644
index 0000000..675a84e
--- /dev/null
+++ b/dataset/freeze/format/__init__.py
@@ -0,0 +1,14 @@
+from dataset.freeze.format.fjson import JSONSerializer
+from dataset.freeze.format.fcsv import CSVSerializer
+from dataset.freeze.format.ftabson import TabsonSerializer
+
+SERIALIZERS = {
+    'json': JSONSerializer,
+    'csv': CSVSerializer,
+    'tabson': TabsonSerializer
+    }
+
+
+def get_serializer(config):
+    serializer = config.get_normalized('format', 'json')
+    return SERIALIZERS.get(serializer)
diff --git a/dataset/freeze/format/common.py b/dataset/freeze/format/common.py
new file mode 100644
index 0000000..bc00af5
--- /dev/null
+++ b/dataset/freeze/format/common.py
@@ -0,0 +1,76 @@
+import os
+import logging
+import re
+import locale
+
+from dataset.util import FreezeException, slug
+
+
+TMPL_KEY = re.compile("{{([^}]*)}}")
+
+OPERATIONS = {
+        'identity': lambda x: x,
+        'lower': lambda x: unicode(x).lower(),
+        'slug': slug
+        }
+
+
+class Serializer(object):
+
+    def __init__(self, engine):
+        self.engine = engine
+        self.config = engine.config
+        self._paths = []
+        self._get_basepath()
+
+    def _get_basepath(self):
+        prefix = self.config.get('prefix')
+        prefix = os.path.abspath(prefix)
+        prefix = os.path.realpath(prefix)
+        self._prefix = prefix
+        filename = self.config.get('filename')
+        if filename is None:
+            raise FreezeException("No 'filename' is specified")
+        self._basepath = os.path.join(prefix, filename)
+
+    def _tmpl(self, data):
+        def repl(m):
+            op, key = 'identity', m.group(1)
+            if ':' in key:
+                op, key = key.split(':', 1)
+            return unicode(OPERATIONS.get(op)(data.get(key, '')))
+        path = TMPL_KEY.sub(repl, self._basepath)
+        enc = locale.getpreferredencoding()
+        return os.path.realpath(path.encode(enc, 'replace'))
+
+    def file_name(self, row):
+        path = self._tmpl(row)
+        if path not in self._paths:
+            if not path.startswith(self._prefix):
+                raise FreezeException("Possible path escape detected.")
+            dn = os.path.dirname(path)
+            if not os.path.isdir(dn):
+                os.makedirs(dn)
+            self._paths.append(path)
+        return path
+
+    @property
+    def mode(self):
+        mode = self.config.get_normalized('mode', 'list')
+        if mode not in ['list', 'item']:
+            raise FreezeException("Invalid mode: %s" % mode)
+        return mode
+
+    @property
+    def wrap(self):
+        return self.config.get_bool('wrap',
+                default=self.mode=='list')
+
+    def serialize(self):
+        self.init()
+        query = self.engine.query()
+        for row in query:
+            self.write(self.file_name(row), row)
+        self.close()
+
+
diff --git a/dataset/freeze/format/fcsv.py b/dataset/freeze/format/fcsv.py
new file mode 100644
index 0000000..00413f6
--- /dev/null
+++ b/dataset/freeze/format/fcsv.py
@@ -0,0 +1,37 @@
+import csv
+from datetime import datetime
+
+from dataset.freeze.format.common import Serializer
+
+
+def value_to_str(value):
+    if isinstance(value, datetime):
+        return value.isoformat()
+    if isinstance(value, unicode):
+        return value.encode('utf-8')
+    if value is None:
+        return ''
+    return value
+
+
+class CSVSerializer(Serializer):
+
+    def init(self):
+        self.handles = {}
+
+    def write(self, path, result):
+        keys = result.keys()
+        if not path in self.handles:
+            fh = open(path, 'wb')
+            writer = csv.writer(fh)
+            writer.writerow([k.encode('utf-8') for k in keys])
+            self.handles[path] = (writer, fh)
+        writer, fh = self.handles[path]
+        values = [value_to_str(result.get(k)) for k in keys]
+        writer.writerow(values)
+
+    def close(self):
+        for (writer, fh) in self.handles.values():
+            fh.close()
+
+
diff --git a/dataset/freeze/format/fjson.py b/dataset/freeze/format/fjson.py
new file mode 100644
index 0000000..446a33e
--- /dev/null
+++ b/dataset/freeze/format/fjson.py
@@ -0,0 +1,45 @@
+import json
+from datetime import datetime
+from collections import defaultdict
+
+from dataset.freeze.format.common import Serializer
+
+
+class JSONEncoder(json.JSONEncoder):
+
+    def default(self, obj):
+        if isinstance(obj, datetime):
+            return obj.isoformat()
+
+
+class JSONSerializer(Serializer):
+
+    def init(self):
+        self.buckets = defaultdict(list)
+
+    def write(self, path, result):
+        self.buckets[path].append(result)
+
+    def wrap(self, result):
+        count = len(result)
+        if self.mode == 'item':
+            result = result[0]
+        if self.wrap:
+            result = {
+                'count': count,
+                'results': result
+                }
+            meta = self.config.get('meta')
+            if meta is not None:
+                result['meta'] = meta
+        return result
+
+    def close(self):
+        for path, result in self.buckets.items():
+            result = self.wrap(result)
+            fh = open(path, 'wb')
+            json.dump(result, fh,
+                    cls=JSONEncoder,
+                    indent=self.config.get_int('indent'))
+            fh.close()
+
diff --git a/dataset/freeze/format/ftabson.py b/dataset/freeze/format/ftabson.py
new file mode 100644
index 0000000..f49aefc
--- /dev/null
+++ b/dataset/freeze/format/ftabson.py
@@ -0,0 +1,24 @@
+from dataset.freeze.format.fjson import JSONSerializer
+
+
+class TabsonSerializer(JSONSerializer):
+
+    def wrap(self, result):
+        fields = []
+        data = []
+        if len(result):
+            keys = result[0].keys()
+            fields = [{'id': k} for k in keys]
+            for row in result:
+                d = [row.get(k) for k in keys]
+                data.append(d)
+        result = {
+            'count': len(result),
+            'fields': fields,
+            'data': data
+            }
+        meta = self.config.get('meta')
+        if meta is not None:
+            result['meta'] = meta
+        return result
+
diff --git a/dataset/util.py b/dataset/util.py
new file mode 100644
index 0000000..339a5fa
--- /dev/null
+++ b/dataset/util.py
@@ -0,0 +1,54 @@
+#coding: utf-8
+import re
+from unicodedata import normalize as ucnorm, category
+
+SLUG_REMOVE = re.compile(r'[,\s\.\(\)/\\;:]*')
+
+class DatasetException(Exception):
+    pass
+
+class FreezeException(DatasetException):
+    pass
+
+
+def normalize(text):
+    """ Simplify a piece of text to generate a more canonical 
+    representation. This involves lowercasing, stripping trailing
+    spaces, removing symbols, diacritical marks (umlauts) and 
+    converting all newlines etc. to single spaces.
+    """
+    if not isinstance(text, unicode):
+        text = unicode(text)
+    text = text.lower()
+    decomposed = ucnorm('NFKD', text)
+    filtered = []
+    for char in decomposed:
+        cat = category(char)
+        if cat.startswith('C'):
+            filtered.append(' ')
+        elif cat.startswith('M'):
+            # marks, such as umlauts
+            continue
+        elif cat.startswith('Z'):
+            # newlines, non-breaking etc.
+            filtered.append(' ')
+        elif cat.startswith('S'):
+            # symbols, such as currency
+            continue
+        else:
+            filtered.append(char)
+    text = u''.join(filtered)
+    while '  ' in text:
+        text = text.replace('  ', ' ')
+    text = text.strip()
+    return ucnorm('NFKC', text)
+
+def slug(text):
+    """ Create a version of a string convenient for use in a URL
+    or file name. """
+    text = normalize(text)
+    text = text.replace(u'ß', 'ss')
+    text = '-'.join(filter(lambda t: len(t), \
+        SLUG_REMOVE.split(text)))
+    return text.lower()
+
diff --git a/setup.py b/setup.py
index 34dedc2..6962a7e 100644
--- a/setup.py
+++ b/setup.py
@@ -23,9 +23,14 @@ setup(
     zip_safe=False,
     install_requires=[
         'sqlalchemy>=0.7',
-        'sqlalchemy-migrate>=0.7'
+        'sqlalchemy-migrate>=0.7',
+        "argparse >= 1.2.1",
+        "PyYAML >= 3.10"
     ],
     tests_require=[],
-    entry_points=\
-    """ """,
+    entry_points={
+        'console_scripts': [
+            'datafreeze = dataset.freeze.app:main',
+        ]
+    }
 )