From 5378208e6d1cfa854d5c12c6f8d60f70bc1e14c3 Mon Sep 17 00:00:00 2001
From: Thiago Curvelo <tcurvelo@gmail.com>
Date: Sun, 20 Oct 2019 19:27:27 -0300
Subject: [PATCH] Wrap DataFrames in order to allow custom styling (closes #76)

---
 src/arche/readers/items.py            |  9 +++++----
 src/arche/tools/dataframe.py          | 15 +++++++++++++++
 tests/tools/test_stylizeddataframe.py | 18 ++++++++++++++++++
 3 files changed, 38 insertions(+), 4 deletions(-)
 create mode 100644 src/arche/tools/dataframe.py
 create mode 100644 tests/tools/test_stylizeddataframe.py

diff --git a/src/arche/readers/items.py b/src/arche/readers/items.py
index 93453a9..4353466 100755
--- a/src/arche/readers/items.py
+++ b/src/arche/readers/items.py
@@ -4,6 +4,7 @@
 
 from arche import SH_URL
 from arche.tools import api
+from arche.tools.dataframe import StylizedDataFrame
 import numpy as np
 import pandas as pd
 from scrapinghub import ScrapinghubClient
@@ -22,11 +23,11 @@ def __len__(self) -> int:
         return len(self.df)
 
     @staticmethod
-    def process_df(df: pd.DataFrame) -> pd.DataFrame:
+    def process_df(df: pd.DataFrame) -> StylizedDataFrame:
         # clean empty objects - mainly lists and dicts, but keep everything else
         df = df.applymap(lambda x: x if x or isinstance(x, numbers.Real) else np.nan)
         Items.categorize(df)
-        return df
+        return StylizedDataFrame(df, copy=True)
 
     @staticmethod
     def categorize(df: pd.DataFrame) -> pd.DataFrame:
@@ -47,7 +48,7 @@ def from_df(cls, df: pd.DataFrame):
 
     @classmethod
     def from_array(cls, iterable: RawItems):
-        return cls(raw=iterable, df=pd.DataFrame(list(iterable)))
+        return cls(raw=iterable, df=StylizedDataFrame(list(iterable)))
 
 
 class CloudItems(Items):
@@ -62,7 +63,7 @@ def __init__(
         self._limit: int = 0
         self.filters = filters
         raw = self.fetch_data()
-        df = pd.DataFrame(list(raw))
+        df = StylizedDataFrame(list(raw))
         df.index = self.format_keys(df["_key"])
         df.index.name = None
         df = df.drop(columns=["_key", "_type"], errors="ignore")
diff --git a/src/arche/tools/dataframe.py b/src/arche/tools/dataframe.py
new file mode 100644
index 0000000..2893307
--- /dev/null
+++ b/src/arche/tools/dataframe.py
@@ -0,0 +1,15 @@
+import re
+
+import pandas as pd
+
+
+class StylizedDataFrame(pd.DataFrame):
+    def _make_urls_clickable(self, val):
+        if isinstance(val, str) and re.search("^https?://", val):
+            return f'<a target="_blank" href="{val}">{val}</a>'
+        else:
+            return val
+
+    def to_html(self, *args, **kwargs):
+        styler = self.style.format(self._make_urls_clickable)
+        return styler.render(*args, **kwargs)
diff --git a/tests/tools/test_stylizeddataframe.py b/tests/tools/test_stylizeddataframe.py
new file mode 100644
index 0000000..d9dc767
--- /dev/null
+++ b/tests/tools/test_stylizeddataframe.py
@@ -0,0 +1,18 @@
+from arche.tools.dataframe import StylizedDataFrame
+import pandas as pd
+
+
+def test_stylized_df_make_urls_clickable():
+    data = {"col1": [1, 2], "col2": ["http://foo.com", "https://bar.com"]}
+    df1 = pd.DataFrame(data)
+    assert "<a href=" not in df1.to_html()
+
+    html = StylizedDataFrame(df1).to_html()
+    assert '<a target="_blank" href="http://foo.com">http://foo.com</a>' in html
+    assert '<a target="_blank" href="https://bar.com">https://bar.com</a>' in html
+
+
+def test_stylized_df_does_not_add_links_if_no_url_found():
+    data = {"col1": [1, 2], "col2": ["foo", "bar"]}
+    html = pd.DataFrame(data).to_html()
+    assert "<a href=" not in html