Merge pull request #59 from Roche/dev

v0.3.3

change_log.md

@@ -1,3 +1,6 @@
+# 0.3.3 (github, pypi and conda 20200424)
+* added capability to set_value_labels to handle subset of dataframes, solves #58
+
 # 0.3.2 (github, pypi and conda 20200422)
 * fixed bug when writing string columns with missing values (#54 and #55)
 

docs/_build/html/.buildinfo

@@ -1,4 +1,4 @@
 # Sphinx build info version 1
 # This file hashes the configuration used when building these files. When it is not found, a full rebuild will be done.
-config: 3c3bb2a4bec2d6479578309a09db4796
+config: e5aad791b428aec4c4ae13c2e09050db
 tags: 645f666f9bcd5a90fca523b33c5a78b7

docs/_build/html/_static/documentation_options.js

@@ -1,6 +1,6 @@
 var DOCUMENTATION_OPTIONS = {
     URL_ROOT: document.getElementById("documentation_options").getAttribute('data-url_root'),
-    VERSION: '0.3.2',
+    VERSION: '0.3.3',
     LANGUAGE: 'None',
     COLLAPSE_INDEX: false,
     BUILDER: 'html',

docs/_build/html/genindex.html

@@ -9,7 +9,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
 
-  <title>Index &mdash; pyreadstat 0.3.2 documentation</title>
+  <title>Index &mdash; pyreadstat 0.3.3 documentation</title>
 
 
 

docs/_build/html/index.html

@@ -8,7 +8,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
 
-  <title>Welcome to pyreadstat’s documentation! &mdash; pyreadstat 0.3.2 documentation</title>
+  <title>Welcome to pyreadstat’s documentation! &mdash; pyreadstat 0.3.3 documentation</title>
 
 
 

docs/_build/html/py-modindex.html

@@ -8,7 +8,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
 
-  <title>Python Module Index &mdash; pyreadstat 0.3.2 documentation</title>
+  <title>Python Module Index &mdash; pyreadstat 0.3.3 documentation</title>
 
 
 

docs/_build/html/search.html

@@ -8,7 +8,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
 
-  <title>Search &mdash; pyreadstat 0.3.2 documentation</title>
+  <title>Search &mdash; pyreadstat 0.3.3 documentation</title>
 
 
 

docs/conf.py

@@ -26,7 +26,7 @@
 # The short X.Y version
 version = ''
 # The full version, including alpha/beta/rc tags
-release = '0.3.2'
+release = '0.3.3'
 
 
 # -- General configuration ---------------------------------------------------

pyreadstat/__init__.py

@@ -20,4 +20,4 @@
 from .pyreadstat import read_file_in_chunks
 from ._readstat_parser import ReadstatError, metadata_container
 
-__version__ = "0.3.2"
+__version__ = "0.3.3"

pyreadstat/_readstat_writer.pyx

@@ -557,8 +557,16 @@ cdef int run_write(df, str filename_path, dst_file_format file_format, str file_
     cdef char *file_labl
 
     cdef list col_names = df.columns.values.tolist()
+
+    for variable_name in col_names:
+        if type(variable_name) != str:
+                raise PyreadstatError("variable name %s is of type %s and it must be str (not starting with numbers!)" % (variable_name, str(type(variable_name))))
+        if not variable_name[0].isalpha():
+            raise PyreadstatError("variable name %s starts with an illegal (non-alphabetic) character" % variable_name)
+
     if file_format == FILE_FORMAT_POR:
         col_names = [x.upper() for x in col_names]
+
     cdef list col_types = get_pandas_column_types(df, missing_user_values)
     cdef int row_count = len(df)
     cdef int col_count = len(col_names)
@@ -618,10 +626,6 @@ cdef int run_write(df, str filename_path, dst_file_format file_format, str file_
             #if file_format == FILE_FORMAT_XPORT and curtype == PYWRITER_DOUBLE:
             #    max_length = 8
             variable_name = col_names[col_indx]
-            if type(variable_name) != str:
-                raise PyreadstatError("variable name %s is of type %s and it must be str (not starting with numbers!)" % (variable_name, str(type(variable_name))))
-            if not variable_name[0].isalpha():
-                raise PyreadstatError("variable name %s starts with an illegal (non-alphabetic) character" % variable_name)
             variable = readstat_add_variable(writer, variable_name.encode("utf-8"), pandas_to_readstat_types[curtype], max_length)
             if curtype in pyrwriter_datetimelike_types:
                 curformat = get_datetimelike_format_for_readstat(file_format, curtype)

pyreadstat/pyreadstat.pyx

@@ -474,9 +474,10 @@ def set_value_labels(dataframe, metadata, formats_as_category=True):
         for var_name, label_name in metadata.variable_to_label.items():
             labels = metadata.value_labels.get(label_name)
             if labels:
-                df_copy[var_name] = df_copy[var_name].apply(lambda x: labels.get(x, x))
-                if formats_as_category:
-                    df_copy[var_name] = df_copy[var_name].astype("category")
+                if var_name in df_copy.columns:
+                    df_copy[var_name] = df_copy[var_name].apply(lambda x: labels.get(x, x))
+                    if formats_as_category:
+                        df_copy[var_name] = df_copy[var_name].astype("category")
 
     return df_copy
 

setup.py

@@ -136,7 +136,7 @@
 
 setup(
     name='pyreadstat',
-    version='0.3.2',
+    version='0.3.3',
     description=short_description,
     author="Otto Fajardo",
     author_email="[email protected]",

tests/test_basic.py

@@ -800,6 +800,18 @@ def test_dta_write_charnan(self):
         df2['object'] = df2['object'].astype(str)
         self.assertTrue(df2.equals(df))
 
+    def test_set_value_labels(self):
+
+        df, meta = pyreadstat.read_sav(os.path.join(self.basic_data_folder, "sample.sav"))
+        df_formatted = pyreadstat.set_value_labels(df, meta, formats_as_category=True)
+        #df.columns = self.df_pandas_formatted.columns
+        self.assertTrue(df_formatted.equals(self.df_pandas_formatted))
+        # partial
+        sub1_raw = df[['myord']]
+        sub1 = pyreadstat.set_value_labels(sub1_raw, meta, formats_as_category=True)
+        sub2 = self.df_pandas_formatted[['myord']]
+        self.assertTrue(sub1.equals(sub2))
+
 if __name__ == '__main__':
 
     import sys