add pdf

2019-10-28 17:23:05 +08:00
parent 9281cb885a
commit 10418167b3
6 changed files with 16 additions and 0 deletions
--- a/pdf/01.png
+++ b/pdf/01.png
--- a/pdf/02.png
+++ b/pdf/02.png
--- a/pdf/03.png
+++ b/pdf/03.png
--- a/pdf/04.png
+++ b/pdf/04.png
--- a/pdf/pdf.py
+++ b/pdf/pdf.py
@@ -0,0 +1,16 @@
+#!/usr/bin/env python3
+
+import pdfplumber
+import pandas as pd
+
+with pdfplumber.open("test.pdf") as pdf:
+    first_page = pdf.pages[0]
+    text = first_page.extract_text()
+    print(text)
+
+    second_page = pdf.pages[1]
+    table = second_page.extract_tables()
+    for t in table:
+        df = pd.DataFrame(t[1:],columns=t[0])
+        print(df)
+
--- a/pdf/test.pdf
+++ b/pdf/test.pdf