Added non-git source puller functionality

Handles non-git source compressed
archives from google drive, dropbox, and any publicly
available web address.

MANIFEST.in

@@ -1,5 +1,6 @@
 include *.md
 include LICENSE
 include setup.cfg
+recursive-include nbgitpuller/plugins *
 recursive-include nbgitpuller/static *
 recursive-include nbgitpuller/templates *

nbgitpuller/handlers.py

@@ -11,6 +11,12 @@
 
 from .pull import GitPuller
 from .version import __version__
+from .hookspecs import NonGitSourceSpec
+from .plugins.zip_puller import ZipSourceGoogleDriveDownloader
+from .plugins.zip_puller import ZipSourceDropBoxDownloader
+from .plugins.zip_puller import ZipSourceWebDownloader
+import pluggy
+import distutils.util
 
 
 class SyncHandler(IPythonHandler):
@@ -38,6 +44,17 @@ def emit(self, data):
         self.write('data: {}\n\n'.format(serialized_data))
         yield self.flush()
 
+    def setup_plugins(self, repo):
+        pm = pluggy.PluginManager("nbgitpuller")
+        pm.add_hookspecs(NonGitSourceSpec)
+        if "drive.google.com" in repo:
+            pm.register(ZipSourceGoogleDriveDownloader())
+        elif "dropbox.com" in repo:
+            pm.register(ZipSourceDropBoxDownloader())
+        else:
+            pm.register(ZipSourceWebDownloader())
+        return pm
+
     @web.authenticated
     @gen.coroutine
     def get(self):
@@ -53,6 +70,7 @@ def get(self):
         try:
             repo = self.get_argument('repo')
             branch = self.get_argument('branch', None)
+            compressed = self.get_argument('compressed', "false")
             depth = self.get_argument('depth', None)
             if depth:
                 depth = int(depth)
@@ -73,6 +91,12 @@ def get(self):
             self.set_header('content-type', 'text/event-stream')
             self.set_header('cache-control', 'no-cache')
 
+            if distutils.util.strtobool(compressed):
+                pm = self.setup_plugins(repo)
+                results = pm.hook.handle_files(repo=repo, repo_parent_dir=repo_parent_dir)[0]
+                repo_dir = repo_parent_dir + results["unzip_dir"]
+                repo = "file://" + results["origin_repo_path"]
+
             gp = GitPuller(repo, repo_dir, branch=branch, depth=depth, parent=self.settings['nbapp'])
 
             q = Queue()
@@ -151,16 +175,19 @@ def get(self):
         repo = self.get_argument('repo')
         branch = self.get_argument('branch', None)
         depth = self.get_argument('depth', None)
+        compressed = self.get_argument('compressed', "false")
         urlPath = self.get_argument('urlpath', None) or \
-                  self.get_argument('urlPath', None)
+            self.get_argument('urlPath', None)
         subPath = self.get_argument('subpath', None) or \
-                  self.get_argument('subPath', '.')
+            self.get_argument('subPath', '.')
         app = self.get_argument('app', app_env)
         parent_reldir = os.getenv('NBGITPULLER_PARENTPATH', '')
         targetpath = self.get_argument('targetpath', None) or \
-                     self.get_argument('targetPath', repo.split('/')[-1])
+            self.get_argument('targetPath', repo.split('/')[-1])
 
-        if urlPath:
+        if distutils.util.strtobool(compressed):
+            path = 'tree/'
+        elif urlPath:
             path = urlPath
         else:
             path = os.path.join(parent_reldir, targetpath, subPath)
@@ -174,7 +201,13 @@ def get(self):
         self.write(
             self.render_template(
                 'status.html',
-                repo=repo, branch=branch, path=path, depth=depth, targetpath=targetpath, version=__version__
+                repo=repo,
+                branch=branch,
+                compressed=compressed,
+                path=path,
+                depth=depth,
+                targetpath=targetpath,
+                version=__version__
             ))
         self.flush()
 

nbgitpuller/hookspecs.py

@@ -0,0 +1,18 @@
+import pluggy
+
+hookspec = pluggy.HookspecMarker("nbgitpuller")
+
+
+class NonGitSourceSpec(object):
+    @hookspec
+    def handle_files(self, repo, repo_parent_dir):
+        """
+         This handles the downloading of non-git source
+         files into the user directory. Once downloaded,
+         the files are merged into a local git repository.
+
+         Once the local git repository is updated(or created
+         the first time), git puller can then handle this
+         directory as it would sources coming from a
+         git repository.
+        """

nbgitpuller/plugins/plugin_helper.py

@@ -0,0 +1,146 @@
+import subprocess
+import os
+import stat
+import logging
+import requests
+from requests_file import FileAdapter
+import shutil
+import re
+
+
+# for large files from Google Drive
+def get_confirm_token(response):
+    for key, value in response.cookies.items():
+        if key.startswith('download_warning'):
+            return value
+    return None
+
+
+# sets up the a local repo that acts like a remote
+def initialize_local_repo(local_repo_path):
+    logging.info(f"Creating local_repo_path: {local_repo_path}")
+    try:
+        os.makedirs(local_repo_path)
+    except OSError:
+        logging.info(f"Directory exists: {local_repo_path}")
+
+    subprocess_helper("git init --bare", local_repo_path)
+
+    # Make our bare repository serveable over dumb HTTP
+    hook_path = os.path.join(local_repo_path, 'hooks', 'post-update')
+    os.rename(
+        os.path.join(local_repo_path, 'hooks', 'post-update.sample'),
+        hook_path
+    )
+    os.chmod(hook_path, os.stat(hook_path).st_mode | stat.S_IEXEC)
+
+
+# local repo cloned from the "remote" which is in user drive
+def clone_local_origin_repo(origin_repo_path, temp_download_repo):
+    logging.info(f"Creating temp_download_repo: {temp_download_repo}")
+    try:
+        os.makedirs(temp_download_repo)
+    except OSError:
+        logging.info(f"Directory exists: {temp_download_repo}")
+
+    cmd = f"git clone file://{origin_repo_path} {temp_download_repo}"
+    subprocess_helper(cmd, temp_download_repo)
+
+
+# this is needed to unarchive various formats(eg. zip, tgz, etc)
+def determine_file_extension(url, response):
+    file_type = response.headers.get('content-type')
+    content_disposition = response.headers.get('content-disposition')
+    ext = None
+    if content_disposition:
+        fname = re.findall("filename\*?=([^;]+)", content_disposition)
+        fname = fname[0].strip().strip('"')
+        ext = fname.split(".")[1]
+    elif file_type and "/zip" in file_type:
+        ext = "zip"
+    else:
+        url = url.split("/")[-1]
+        if "?" in url:
+            url = url[0:url.find('?')]
+        if "." in url:
+            ext = url.split(".")[1]
+
+    if not ext:
+        m = f"Could not determine the file extension for unarchiving: {url}"
+        raise Exception(m)
+    return ext
+
+
+# the downloaded content is in the response -- unarchive and save to the disk
+def save_response_content(url, response, temp_download_repo):
+    try:
+        ext = determine_file_extension(url, response)
+        CHUNK_SIZE = 32768
+        temp_download_file = f"{temp_download_repo}/download.{ext}"
+        with open(temp_download_file, "wb") as f:
+            for chunk in response.iter_content(CHUNK_SIZE):
+                # filter out keep-alive new chunks
+                if chunk:
+                    f.write(chunk)
+
+        shutil.unpack_archive(temp_download_file, temp_download_repo)
+
+        os.remove(temp_download_file)
+    except Exception as e:
+        m = f"Problem handling file download: {str(e)}"
+        raise Exception(m)
+
+
+# grab archive file from url
+def fetch_files(url, id=-1):
+    session = requests.Session()
+    session.mount('file://', FileAdapter())  # add adapter for pytests
+    response = session.get(url, params={'id': id}, stream=True)
+    token = get_confirm_token(response)
+    if token:
+        params = {'id': id, 'confirm': token}
+        response = session.get(url, params=params, stream=True)
+
+    return response
+
+
+# this drive the file handling -- called from zip_puller by all the
+# handle_files implementations for GoogleDrive, Dropbox, and standard
+# Web url
+def handle_files_helper(args):
+    try:
+        origin_repo = args["repo_parent_dir"] + args["origin_dir"]
+        temp_download_repo = args["repo_parent_dir"] + args["download_dir"]
+        if os.path.exists(temp_download_repo):
+            shutil.rmtree(temp_download_repo)
+
+        if not os.path.exists(origin_repo):
+            initialize_local_repo(origin_repo)
+
+        clone_local_origin_repo(origin_repo, temp_download_repo)
+        save_response_content(args["repo"], args["response"], temp_download_repo)
+        subprocess_helper("git add .", temp_download_repo)
+        subprocess_helper("git -c [email protected] -c user.name=nbgitpuller commit -m test --allow-empty", temp_download_repo)
+        subprocess_helper("git push origin master", temp_download_repo)
+        unzipped_dirs = os.listdir(temp_download_repo)
+
+        dir_names = list(filter(lambda dir: ".git" not in dir, unzipped_dirs))
+        return {"unzip_dir": dir_names[0], "origin_repo_path": origin_repo}
+    except Exception as e:
+        logging.exception(e)
+        raise ValueError(e)
+
+
+# executes git commands for us
+def subprocess_helper(cmd, cwd):
+    try:
+        subprocess.run(
+            cmd.split(" "),
+            capture_output=True,
+            text=True,
+            check=True,
+            cwd=cwd
+        )
+    except Exception:
+        m = f"Problem executing git command: {cmd}"
+        raise Exception(m)

nbgitpuller/plugins/zip_puller.py

@@ -0,0 +1,67 @@
+from .plugin_helper import fetch_files
+from .plugin_helper import handle_files_helper
+import pluggy
+
+hookimpl = pluggy.HookimplMarker("nbgitpuller")
+TEMP_DOWNLOAD_REPO_DIR = ".temp_download_repo"
+CACHED_ORIGIN_NON_GIT_REPO = ".origin_non_git_sources"
+
+
+# handles standard web addresses(not google drive or dropbox)
+class ZipSourceWebDownloader(object):
+    @hookimpl
+    # repo --> this is download url
+    # repo_parent_dir --> where we will create the repo
+    def handle_files(self, repo, repo_parent_dir):
+        response = fetch_files(repo)
+        args = {
+            "repo": repo,
+            "repo_parent_dir": repo_parent_dir,
+            "response": response,
+            "origin_dir": CACHED_ORIGIN_NON_GIT_REPO,
+            "download_dir": TEMP_DOWNLOAD_REPO_DIR
+        }
+        return handle_files_helper(args)
+
+
+# handles downloads from google drive
+class ZipSourceGoogleDriveDownloader(object):
+    def __init__(self):
+        self.DOWNLOAD_URL = "https://docs.google.com/uc?export=download"
+
+    def get_id(self, repo):
+        start_id_index = repo.index("d/") + 2
+        end_id_index = repo.index("/view")
+        return repo[start_id_index:end_id_index]
+
+    # repo --> this is download url - it has the id
+    # repo_parent_dir --> where we will create the repo
+    @hookimpl
+    def handle_files(self, repo, repo_parent_dir):
+        response = fetch_files(self.DOWNLOAD_URL, self.get_id(repo))
+        args = {
+            "repo": repo,
+            "repo_parent_dir": repo_parent_dir,
+            "response": response,
+            "origin_dir": CACHED_ORIGIN_NON_GIT_REPO,
+            "download_dir": TEMP_DOWNLOAD_REPO_DIR
+        }
+        return handle_files_helper(args)
+
+
+# handles downloads from DropBox
+class ZipSourceDropBoxDownloader(object):
+    # repo --> this is download url
+    # repo_parent_dir --> where we will create the repo
+    @hookimpl
+    def handle_files(self, repo, repo_parent_dir):
+        repo = repo.replace("dl=0", "dl=1")  # download set to 1 for dropbox
+        response = fetch_files(repo)
+        args = {
+            "repo": repo,
+            "repo_parent_dir": repo_parent_dir,
+            "response": response,
+            "origin_dir": CACHED_ORIGIN_NON_GIT_REPO,
+            "download_dir": TEMP_DOWNLOAD_REPO_DIR
+        }
+        return handle_files_helper(args)

nbgitpuller/static/index.js

@@ -12,12 +12,13 @@ require([
 
     Terminal.applyAddon(fit);
 
-    function GitSync(baseUrl, repo, branch, depth, targetpath, path) {
+    function GitSync(baseUrl, repo, branch, depth, compressed, targetpath, path) {
         // Class that talks to the API backend & emits events as appropriate
         this.baseUrl = baseUrl;
         this.repo = repo;
         this.branch = branch;
         this.depth = depth;
+        this.compressed = compressed;
         this.targetpath = targetpath;
         this.redirectUrl = baseUrl + path;
 
@@ -52,6 +53,9 @@ require([
         if (typeof this.branch !== 'undefined' && this.branch != undefined) {
             syncUrlParams['branch'] = this.branch;
         }
+        if (typeof this.compressed !== 'undefined' && this.compressed != undefined) {
+            syncUrlParams['compressed'] = this.compressed;
+        }
         var syncUrl = this.baseUrl + 'git-pull/api?' + $.param(syncUrlParams);
 
         this.eventSource = new EventSource(syncUrl);
@@ -133,6 +137,7 @@ require([
         utils.get_body_data('repo'),
         utils.get_body_data('branch'),
         utils.get_body_data('depth'),
+        utils.get_body_data('compressed'),
         utils.get_body_data('targetpath'),
         utils.get_body_data('path')
     );

nbgitpuller/templates/status.html

@@ -7,6 +7,7 @@
 data-path="{{ path | urlencode }}"
 {% if branch %}data-branch="{{ branch | urlencode }}"{% endif %}
 {% if depth %}data-depth="{{ depth | urlencode }}"{% endif %}
+{% if compressed %}data-compressed="{{ compressed | urlencode }}"{% endif %}
 data-targetpath="{{ targetpath | urlencode }}"
 {% endblock %}
 

setup.py

@@ -21,7 +21,7 @@
     packages=find_packages(),
     include_package_data=True,
     platforms='any',
-    install_requires=['notebook>=5.5.0', 'tornado'],
+    install_requires=['notebook>=5.5.0', 'tornado', 'requests', 'requests-file'],
     data_files=[
         ('etc/jupyter/jupyter_notebook_config.d', ['nbgitpuller/etc/nbgitpuller.json'])
     ],