From 657371acca235235bbb852014b594ad965cc19d9 Mon Sep 17 00:00:00 2001
From: sarahmccuan <35535105+sarahmccuan@users.noreply.github.com>
Date: Sat, 15 Jun 2024 19:57:31 -0500
Subject: [PATCH] add a dockerfile so this isn't so annoying to build from
 scratch (#19)

---
 Dockerfile           | 20 ++++++++++++++++++++
 build-html.py        |  2 +-
 requirements.txt     |  4 ++++
 scripts/normalize.py | 16 ++++++++--------
 scripts/validate.py  |  2 +-
 5 files changed, 34 insertions(+), 10 deletions(-)
 create mode 100644 Dockerfile
 create mode 100644 requirements.txt

diff --git a/Dockerfile b/Dockerfile
new file mode 100644
index 0000000..9d0ca3a
--- /dev/null
+++ b/Dockerfile
@@ -0,0 +1,20 @@
+FROM python:3.8-slim 
+
+WORKDIR /usr/src/app
+
+COPY requirements.txt ./
+
+RUN pip install --no-cache-dir --upgrade pip \
+  && pip install --no-cache-dir -r requirements.txt
+
+COPY . .
+
+RUN ["python", "scripts/normalize.py"]
+
+RUN ["python", "scripts/validate.py", ">", "test_results.txt"]
+
+RUN ["python", "build-html.py"]
+
+WORKDIR /usr/src/app/docs
+
+CMD ["python", "-m", "http.server"]
\ No newline at end of file
diff --git a/build-html.py b/build-html.py
index dd11d43..0b66511 100644
--- a/build-html.py
+++ b/build-html.py
@@ -84,5 +84,5 @@ def wrap_title(title: str, id = ''):
 print(body)
 html = '\n'.join([HEADER, body, FOOTER])
 
-with open('docs/index.html', 'w', encoding="utf-8") as f:
+with open('docs/greekwar.html', 'w', encoding="utf-8") as f:
     f.write(html)
diff --git a/requirements.txt b/requirements.txt
new file mode 100644
index 0000000..2e4b7fd
--- /dev/null
+++ b/requirements.txt
@@ -0,0 +1,4 @@
+text_validator
+pathlib
+greek-normalisation
+markdown
diff --git a/scripts/normalize.py b/scripts/normalize.py
index 6af028c..3f4ddcb 100644
--- a/scripts/normalize.py
+++ b/scripts/normalize.py
@@ -15,14 +15,14 @@ def normalize_file(filename):
         print('Error: ' + str(sys.exc_info()[0]))
 
 # relative path logic
-chambers_path = Path(__file__).parent.parent
-files_dir = (chambers_path / 'text\\').resolve()
-files_list = glob.glob(str(files_dir) + '*.txt')
+# chambers_path = Path(__file__).parent.parent
+# files_dir = (chambers_path / 'text\\').resolve()
+# files_list = glob.glob(str(files_dir) + '*.txt')
 
-# normalize_file("C:/Users/Sarah McCuan/Documents/projects/ChambersGreekWar/drafts/chambers_ocr.md")
-normalize_file("C:/Users/Sarah McCuan/Documents/projects/ChambersGreekWar/drafts/greek_english_vocab.md")
+# instead of using the list above and potentially messing up in-progress files, 
+# it's easier to just individually specify what we want normalized: 
 
-# for file in files_list:
-#     normalize_file(file)
+files_list = ['text/chambers_w_headers.txt', 'text/chambers.txt', 'docs/greek_english_vocab.html', 'docs/greekwar.html', 'docs/index.html']
 
-# print(validate((chambers_path / 'text-validator.toml').resolve(), files_list))
+for file in files_list:
+    normalize_file(file)
\ No newline at end of file
diff --git a/scripts/validate.py b/scripts/validate.py
index ec15f8f..a57cb18 100644
--- a/scripts/validate.py
+++ b/scripts/validate.py
@@ -1,3 +1,3 @@
 from text_validator.main import validate
 
-validate("C:/Users/Sarah McCuan/Documents/projects/ChambersGreekWar/text-validator.toml", ["C:/Users/Sarah McCuan/Documents/projects/ChambersGreekWar/text/01.txt"])
\ No newline at end of file
+validate("text-validator.toml", ["text/chambers_w_headers.txt"])
\ No newline at end of file