Commands: added SearchHexPackages and OpenHexDocs.

Author: azizk

README.md

@@ -83,6 +83,8 @@ Use the default shortcut `Alt+Shift+F` or the palette command `Mix Format: File`
 ## Palette commands
 
 - `ElixirSyntax: Settings`
+- `ElixirSyntax: Open Hex Docs`
+- `ElixirSyntax: Search Hex Packages`
 - `Mix Test: Settings`
 - `Mix Test: All`
 - `Mix Test: File`

commands/Default.sublime-commands

@@ -3,6 +3,8 @@
     "base_file": "${packages}/ElixirSyntax/settings/ElixirSyntax.sublime-settings",
     "default": "{\n  $0\n}\n"
   } },
+  { "caption": "ElixirSyntax: Open Hex Docs", "command": "open_hex_docs" },
+  { "caption": "ElixirSyntax: Search Hex Packages", "command": "search_hex_packages" },
   { "caption": "Mix Test: Settings", "command": "mix_test_settings" },
   { "caption": "Mix Test: All", "command": "mix_test" },
   { "caption": "Mix Test: File", "command": "mix_test_file" },

commands/__init__.py

@@ -1,2 +1,3 @@
+from .hex_packages import *
 from .mix_test import *
 from .mix_format import *

commands/hex_packages.py

@@ -0,0 +1,210 @@
+import sublime
+import sublime_plugin
+import re
+import webbrowser
+
+from pathlib import Path
+from urllib import request
+from urllib.error import HTTPError
+from datetime import datetime
+from .utils import *
+
+__author__ = 'Aziz Köksal'
+__email__ = '[email protected]'
+__status__ = 'Production'
+
+HEXDOCS_URL = 'https://hexdocs.pm'
+HEX_URL = 'https://hex.pm'
+ELIXIR_CORE_APP_NAMES = ['eex', 'elixir', 'ex_unit', 'hex', 'iex', 'logger', 'mix']
+name_lastmod_rx = r'hexdocs.pm/([^/]+)/[\s\S]+?<lastmod>([^<]+)</lastmod>'
+PROJECT_MAX_AGE_DAYS = 365
+
+class SearchHexPackagesCommand(sublime_plugin.WindowCommand):
+  def description(self):
+    return 'Searches hex.pm and shows the results.'
+
+  def run(self, **kwargs):
+    query = (kwargs.get('query') or '').strip()
+
+    if query:
+      print_status_msg('Searching hex.pm for %r' % query)
+      sublime.set_timeout_async(lambda: search_hex_pm(self.window, query))
+
+  def input(self, _args):
+    class QueryInputHandler(sublime_plugin.TextInputHandler):
+      def placeholder(self): return 'Search hex.pm'
+      def validate(self, text): return text.strip() != ''
+
+    return QueryInputHandler()
+
+class OpenHexDocsCommand(sublime_plugin.WindowCommand):
+  def description(self):
+    return 'Finds and opens hex documentation in the browser.'
+
+  def run(self, **_kwargs):
+    cache_dir = Path(sublime.cache_path(), 'ElixirSyntax')
+    cache_dir.exists() or cache_dir.mkdir(parents=True)
+
+    cached_sitemap_json_path = Path(cache_dir, 'hexdocs.sitemap.json')
+
+    sitemap_dict = {}
+    sitemap_url = HEXDOCS_URL + '/sitemap.xml'
+
+    if cached_sitemap_json_path.exists():
+      sitemap_dict = load_json_file(cached_sitemap_json_path)
+      etag = sitemap_dict['etag']
+
+      def refresh_sitemap():
+        try:
+          resp = request.urlopen(request.Request(sitemap_url, headers={'If-None-Match': etag}))
+          sitemap_dict = fetch_parse_and_save_sitemap(resp, cached_sitemap_json_path)
+          show_hexdocs_list(self.window, sitemap_dict.get('projects', []))
+        except HTTPError as e:
+          e.code == 304 or print_status_msg('Error: %s' % e)
+
+      sublime.set_timeout_async(refresh_sitemap)
+
+      show_hexdocs_list(self.window, sitemap_dict.get('projects', []))
+    else:
+      print_status_msg('Downloading %r' % sitemap_url)
+
+      def fetch_sitemap():
+        try:
+          resp = request.urlopen(sitemap_url)
+          sitemap_dict = fetch_parse_and_save_sitemap(resp, cached_sitemap_json_path)
+          show_hexdocs_list(self.window, sitemap_dict.get('projects', []))
+        except HTTPError as e:
+          print_status_msg('Error: could not fetch %r (status=#%s)' % (sitemap_url, resp.code))
+
+      sublime.set_timeout_async(fetch_sitemap)
+
+def search_hex_pm(window, query, **kwargs):
+  """ Searches hex.pm and shows the results in a quick panel overlay. """
+  page = kwargs.get('page')
+  page_param = page and ['page=%s' % page] or []
+  query = query and ''.join('%%%x' % ord(c) if c in '#&/?' else c for c in query)
+  get_params = '&'.join(['search=%s' % query, 'sort=recent_downloads'] + page_param)
+  query_url = HEX_URL + '/packages?' + get_params
+  resp = request.urlopen(query_url)
+  results_html = resp.read().decode('utf-8')
+
+  package_list_match = re.search(r'<div class="package-list">([\s\S]+?)\n</div>', results_html)
+  page_match = re.search(r'<li class="active">[\s\S]+?</li>[\s\S]+?\bpage=(\d+)', results_html)
+  next_page = page_match and int(page_match.group(1))
+  total_count_match = re.search(r'packages of (\d+) total', results_html)
+  total_packages_count = total_count_match and total_count_match.group(1)
+
+  if not package_list_match:
+    has_no_results = 'no-results' in results_html
+
+    msg = [
+        'could not find div.package-list in the results HTML.',
+        'no results found for %r on hex.pm!' % query
+      ][has_no_results]
+
+    if has_no_results:
+      overlay_args = {'overlay': 'command_palette', 'command': 'search_hex_packages'}
+      window.run_command('show_overlay', overlay_args)
+      window.run_command('insert', {'characters': query})
+
+    print_status_msg('Error: ' + msg)
+    return
+
+  package_matches = re.findall(r'''(?xi)
+      <span\sclass="download-count"> (.+?) </span> [\s\S]*?
+      <span.+?> total\sdownloads:\s (.+?) </span> [\s\S]*?
+      <a.+?> (.+?) </a> [\s\S]*?
+      <span.+?> (.+?) </span> [\s\S]*?
+      <p> ([^<]*) </p>
+      ''',
+      package_list_match.group(1)
+    )
+
+  previous_results = kwargs.get('previous_results') or []
+
+  results = previous_results + [
+      {'name': m[2], 'desc': m[4], 'version': m[3], 'recent_dls': m[0], 'total_dls': m[1],
+       'url': HEX_URL + '/packages/' + m[2]}
+      for m in package_matches
+    ]
+
+  selectable_results = results + [
+      {'label': 'Open search query in browser', 'url': query_url, 'desc': 'Terms: %s' % query},
+    ] + (
+      next_page and [{
+        'label': 'Load page %d' % next_page,
+        'page': next_page,
+        'desc': 'Total packages found: %s' % (total_packages_count or 'unknown')
+      }] or []
+    )
+
+  def on_select(i):
+    if i >= 0:
+      result = selectable_results[i]
+      if result.get('page'):
+        print_status_msg('Loading page %d on hex.pm for %r' % (next_page, query))
+        cb = lambda: search_hex_pm(window, query, page=next_page, previous_results=results)
+        sublime.set_timeout_async(cb)
+      else:
+        webbrowser.open_new_tab(result['url'])
+
+  placeholder = 'Open a project in the web browser.'
+  selected_index = len(previous_results) if previous_results else -1
+
+  result_items = [
+      sublime.QuickPanelItem(
+          trigger=result.get('label') or '%s v%s' % (result['name'], result['version']),
+          details=result.get('desc') or '',
+          annotation=result.get('recent_dls') \
+            and '%s recent / %s total downloads' % (result['recent_dls'], result['total_dls']) \
+            or '',
+          kind=result.get('recent_dls') and sublime.KIND_NAVIGATION or sublime.KIND_AMBIGUOUS
+        )
+      for result in selectable_results
+    ]
+
+  window.show_quick_panel(result_items, on_select,
+      placeholder=placeholder, selected_index=selected_index
+    )
+
+def fetch_parse_and_save_sitemap(resp, cached_sitemap_json_path):
+  """ Fetches, parses and saves the sitemap items in a JSON file. """
+  etag = next(
+      (value for (header, value) in resp.headers.items() if header.lower() == 'etag'), None
+    )
+
+  sitemap_xml = resp.read().decode('utf-8')
+  elixir_core_projects = [(name, None) for name in ELIXIR_CORE_APP_NAMES]
+  hexdocs_projects = re.findall(name_lastmod_rx, sitemap_xml)
+  young_projects, old_projects, now = [], [], datetime.now()
+
+  for name, date in hexdocs_projects:
+    parsed_date = datetime.strptime(date[:10], '%Y-%m-%d')
+    younger_than_x_days = (now - parsed_date).days <= PROJECT_MAX_AGE_DAYS
+    (young_projects if younger_than_x_days else old_projects).append((name, date))
+
+  projects = sorted(young_projects + elixir_core_projects) + old_projects
+  projects = [{'name': name, 'lastmod': lastmod} for (name, lastmod) in projects]
+  sitemap_dict = {'projects': projects, 'etag': etag}
+  save_json_file(cached_sitemap_json_path, sitemap_dict)
+
+  return sitemap_dict
+
+def show_hexdocs_list(window, projects):
+  """ Shows the hexdocs projects in a quick panel overlay. """
+  project_items = [
+      sublime.QuickPanelItem(
+          trigger=project['name'],
+          details=project['lastmod'] \
+            and 'Last modified: %s' % project['lastmod'][:-4].replace('T', ' ') \
+            or '',
+          kind=sublime.KIND_NAVIGATION
+        )
+      for project in projects
+    ]
+
+  def on_select(i):
+    i >= 0 and webbrowser.open_new_tab(HEXDOCS_URL + '/' + projects[i]['name'])
+
+  placeholder = 'Open a project\'s documentation in the web browser.'
+  window.show_quick_panel(project_items, on_select, placeholder=placeholder)