Merge pull request galaxyproject#10175 from mvdbeek/ngram_by_default

Use tool_enable_ngram_search by default, whitespace search fix

doc/source/admin/galaxy_options.rst

@@ -1556,6 +1556,21 @@
 :Type: str
 
 
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+``trs_servers_config_file``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+:Description:
+    Allow import of workflows from the TRS servers configured in the
+    specified YAML or JSON file. The file should be a list with 'id',
+    'label', and 'api_url' for each entry. Optionally, 'link_url' and
+    'doc' may be be specified as well for each entry.
+    If this is null (the default), a simple configuration containing
+    just Dockstore will be used.
+:Default: ``trs_servers_conf.yml``
+:Type: str
+
+
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 ``user_preferences_extra_conf_path``
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@@ -2699,7 +2714,7 @@
     Enable/ disable Ngram-search for tools. It makes tool search
     results tolerant for spelling mistakes in the query by dividing
     the query into multiple ngrams and search for each ngram
-:Default: ``false``
+:Default: ``true``
 :Type: bool
 
 
@@ -3296,6 +3311,47 @@
 :Type: bool
 
 
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+``simplified_workflow_run_ui``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+:Description:
+    If set to 'off' by default, always use the traditional workflow
+    form that renders all steps in the GUI and serializes the tool
+    state of all steps during invocation. Set to 'prefer' to default
+    to a simplified workflow UI that only renders the inputs if
+    possible (the workflow must have no disconnected runtime inputs
+    and not replacement parameters within tool steps). In the future
+    'force' may be added an option for Galaskio-style servers that
+    should only render simplified workflows.
+:Default: ``prefer``
+:Type: str
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+``simplified_workflow_run_ui_target_history``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+:Description:
+    When the simplified workflow run form is rendered, should the
+    invocation outputs be sent to the 'current' history or a 'new'
+    history.
+:Default: ``current``
+:Type: str
+
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+``simplified_workflow_run_ui_job_cache``
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+:Description:
+    When the simplified workflow run form is rendered, should the
+    invocation use job caching. This isn't a boolean so an option for
+    'show-selection' can be added later.
+:Default: ``off``
+:Type: str
+
+
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
 ``myexperiment_target_url``
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~

lib/galaxy/config/sample/galaxy.yml.sample

@@ -834,6 +834,14 @@ galaxy:
   # format string as specified by ISO 8601 international   standard).
   #pretty_datetime_format: $locale (UTC)
 
+  # Allow import of workflows from the TRS servers configured in the
+  # specified YAML or JSON file. The file should be a list with 'id',
+  # 'label', and 'api_url' for each entry. Optionally, 'link_url' and
+  # 'doc' may be be specified as well for each entry.
+  # If this is null (the default), a simple configuration containing
+  # just Dockstore will be used.
+  #trs_servers_config_file: trs_servers_conf.yml
+
   # Location of the configuration file containing extra user
   # preferences.
   # The value of this option will be resolved with respect to
@@ -1346,7 +1354,7 @@ galaxy:
   # Enable/ disable Ngram-search for tools. It makes tool search results
   # tolerant for spelling mistakes in the query by dividing the query
   # into multiple ngrams and search for each ngram
-  #tool_enable_ngram_search: false
+  #tool_enable_ngram_search: true
 
   # Set minimum size of ngrams
   #tool_ngram_minsize: 3
@@ -1628,6 +1636,26 @@ galaxy:
   # workflow.
   #enable_unique_workflow_defaults: false
 
+  # If set to 'off' by default, always use the traditional workflow form
+  # that renders all steps in the GUI and serializes the tool state of
+  # all steps during invocation. Set to 'prefer' to default to a
+  # simplified workflow UI that only renders the inputs if possible (the
+  # workflow must have no disconnected runtime inputs and not
+  # replacement parameters within tool steps). In the future 'force' may
+  # be added an option for Galaskio-style servers that should only
+  # render simplified workflows.
+  #simplified_workflow_run_ui: prefer
+
+  # When the simplified workflow run form is rendered, should the
+  # invocation outputs be sent to the 'current' history or a 'new'
+  # history.
+  #simplified_workflow_run_ui_target_history: current
+
+  # When the simplified workflow run form is rendered, should the
+  # invocation use job caching. This isn't a boolean so an option for
+  # 'show-selection' can be added later.
+  #simplified_workflow_run_ui_job_cache: 'off'
+
   # The URL to the myExperiment instance being used (omit scheme but
   # include port).
   #myexperiment_target_url: www.myexperiment.org:80

lib/galaxy/tools/__init__.py

@@ -603,6 +603,13 @@ def is_latest_version(self):
         tool_versions = self.tool_versions
         return not tool_versions or self.version == self.tool_versions[-1]
 
+    @property
+    def latest_version(self):
+        if self.is_latest_version:
+            return self
+        else:
+            return self.app.tool_cache.get_tool_by_id(self.lineage.get_versions()[-1].id)
+
     @property
     def is_datatype_converter(self):
         return self in self.app.datatypes_registry.converter_tools

lib/galaxy/tools/search/__init__.py

@@ -82,12 +82,31 @@ def build_index(self, tool_cache, index_help=True):
             # Index ocasionally contains empty stored fields
             indexed_tool_ids = {f['id'] for f in reader.all_stored_fields() if f}
         tool_ids_to_remove = (indexed_tool_ids - set(tool_cache._tool_paths_by_id.keys())).union(tool_cache._removed_tool_ids)
+        for indexed_tool_id in indexed_tool_ids:
+            indexed_tool = tool_cache.get_tool_by_id(indexed_tool_id)
+            if not indexed_tool:
+                tool_ids_to_remove.add(indexed_tool_id)
+                continue
+            if not indexed_tool.is_latest_version and not indexed_tool.latest_version.hidden:
+                tool_ids_to_remove.add(indexed_tool_id)
         with AsyncWriter(self.index) as writer:
             for tool_id in tool_ids_to_remove:
                 writer.delete_by_term('id', tool_id)
             for tool_id in tool_cache._new_tool_ids - indexed_tool_ids:
                 tool = tool_cache.get_tool_by_id(tool_id)
                 if tool and tool.is_latest_version:
+                    if tool.hidden:
+                        # we check if there is an older tool we can return
+                        if tool.lineage:
+                            for tool_version in reversed(tool.lineage.get_versions()):
+                                tool = tool_cache.get_tool_by_id(tool_version.id)
+                                if tool and not tool.hidden:
+                                    tool_id = tool.id
+                                    break
+                            else:
+                                continue
+                        else:
+                            continue
                     add_doc_kwds = self._create_doc(tool_id=tool_id, tool=tool, index_help=index_help)
                     writer.update_document(**add_doc_kwds)
         log.debug("Toolbox index finished %s", execution_timer)
@@ -154,13 +173,11 @@ def search(self, q, tool_name_boost, tool_section_boost,
         og = OrGroup.factory(0.9)
         self.parser = MultifieldParser(['name', 'description', 'section', 'help', 'labels', 'stub'], schema=self.schema, group=og)
         cleaned_query = q.lower()
-        # Replace hyphens, since they are wildcards in Whoosh causing false positives
-        if cleaned_query.find('-') != -1:
-            cleaned_query = (' ').join(token.text for token in self.rex(to_unicode(cleaned_query)))
         if tool_enable_ngram_search is True:
             rval = self._search_ngrams(cleaned_query, tool_ngram_minsize, tool_ngram_maxsize, tool_search_limit)
             return rval
         else:
+            cleaned_query = ' '.join(token.text for token in self.rex(cleaned_query))
             # Use asterisk Whoosh wildcard so e.g. 'bow' easily matches 'bowtie'
             parsed_query = self.parser.parse(cleaned_query + '*')
             hits = self.searcher.search(parsed_query, limit=float(tool_search_limit), sortedby='')

lib/galaxy/webapps/galaxy/config_schema.yml

@@ -1999,7 +1999,7 @@ mapping:
 
       tool_enable_ngram_search:
         type: bool
-        default: false
+        default: true
         required: false
         reloadable: true
         desc: |