Merge branch 'main' into feature_localtemp

johanneskoester · web-flow · commit 2baf2cdc6986 · 2025-03-14T11:43:31.000+01:00
diff --git a/apidocs/conf.py b/apidocs/conf.py
@@ -22,7 +22,7 @@
 # If extensions (or modules to document with autodoc) are in another directory,
 # add these directories to sys.path here. If the directory is relative to the
 # documentation root, use os.path.abspath to make it absolute, like shown here.
-sys.path.insert(0, os.path.abspath("../"))
+sys.path.insert(0, os.path.abspath("../src/"))
 
 # -- General configuration ------------------------------------------------
 
diff --git a/docs/conf.py b/docs/conf.py
@@ -22,8 +22,7 @@
 # If extensions (or modules to document with autodoc) are in another directory,
 # add these directories to sys.path here. If the directory is relative to the
 # documentation root, use os.path.abspath to make it absolute, like shown here.
-sys.path.insert(0, os.path.abspath("../"))
-
+sys.path.insert(0, os.path.abspath("../src/"))
 # -- General configuration ------------------------------------------------
 
 # If your documentation needs a minimal Sphinx version, state it here.
diff --git a/src/snakemake/cli.py b/src/snakemake/cli.py
@@ -68,6 +68,7 @@
     StorageSettings,
     WorkflowSettings,
     StrictDagEvaluation,
+    PrintDag,
 )
 from snakemake.target_jobs import parse_target_jobs_cli_args
 from snakemake.utils import available_cpu_count, update_config
@@ -1045,20 +1046,26 @@ def get_argument_parser(profiles=None):
         action="store_true",
         help="Show available target rules in given Snakefile.",
     )
-    group_utils.add_argument(
+    group_exec.add_argument(
         "--dag",
-        action="store_true",
+        nargs="?",
+        choices=PrintDag.choices(),
+        const=str(PrintDag.DOT),
+        default=None,
         help="Do not execute anything and print the directed "
-        "acyclic graph of jobs in the dot language. Recommended "
+        "acyclic graph of jobs in the dot language or in mermaid-js. Recommended "
         "use on Unix systems: snakemake `--dag | dot | display`. "
         "Note print statements in your Snakefile may interfere "
         "with visualization.",
     )
     group_utils.add_argument(
         "--rulegraph",
-        action="store_true",
+        nargs="?",
+        choices=PrintDag.choices(),
+        const=str(PrintDag.DOT),
+        default=None,
         help="Do not execute anything and print the dependency graph "
-        "of rules in the dot language. This will be less "
+        "of rules in the dot language or in mermaid-js. This will be less "
         "crowded than above DAG of jobs, but also show less information. "
         "Note that each rule is displayed once, hence the displayed graph will be "
         "cyclic if a rule appears in several steps of the workflow. "
@@ -2025,6 +2032,13 @@ def args_to_api(args, parser):
                 elif args.print_compilation:
                     workflow_api.print_compilation()
                 else:
+
+                    print_dag_as = None
+                    if args.dag:
+                        print_dag_as = args.dag
+                    elif args.rulegraph:
+                        print_dag_as = args.rulegraph
+
                     dag_api = workflow_api.dag(
                         dag_settings=DAGSettings(
                             targets=args.targets,
@@ -2042,6 +2056,7 @@ def args_to_api(args, parser):
                             max_inventory_wait_time=args.max_inventory_time,
                             max_checksum_file_size=args.max_checksum_file_size,
                             strict_evaluation=args.strict_dag_evaluation,
+                            print_dag_as=print_dag_as,
                         ),
                     )
 
diff --git a/src/snakemake/dag.py b/src/snakemake/dag.py
@@ -55,6 +55,7 @@
     WorkflowError,
     print_exception_warning,
 )
+from snakemake.settings.types import PrintDag
 from snakemake.io import (
     _IOFile,
     PeriodicityDetector,
@@ -2382,7 +2383,10 @@ def rule_dot(self):
         graph = defaultdict(set)
         for job in self.jobs:
             graph[job.rule].update(dep.rule for dep in self._dependencies[job])
-        return self._dot(graph)
+        if self.workflow.dag_settings.print_dag_as == str(PrintDag.DOT):
+            return self._dot(graph)
+        elif self.workflow.dag_settings.print_dag_as == str(PrintDag.MERMAID_JS):
+            return self._mermaid_js(graph)
 
     def dot(self):
         def node2style(job):
@@ -2407,6 +2411,70 @@ def format_wildcard(wildcard):
             dag, node2rule=node2rule, node2style=node2style, node2label=node2label
         )
 
+    def mermaid_js(self):
+        def node2style(job):
+            if not self.needrun(job):
+                return ",stroke-dasharray: 5 5"
+            return ""
+
+        def format_wildcard(wildcard):
+            name, value = wildcard
+            return f"{name}: {value}"
+
+        node2label = lambda job: " - ".join(
+            chain(
+                [job.rule.name], sorted(map(format_wildcard, self.new_wildcards(job)))
+            )
+        )
+
+        dag = {job: self._dependencies[job] for job in self.jobs}
+
+        return self._mermaid_js(dag, node2style=node2style, node2label=node2label)
+
+    def _mermaid_js(
+        self, graph, node2style=lambda node: "", node2label=lambda node: node
+    ):
+        def hsv_to_htmlhexrgb(h, s, v):
+            import colorsys
+
+            hex_r, hex_g, hex_b = (round(255 * x) for x in colorsys.hsv_to_rgb(h, s, v))
+            return "#{hex_r:0>2X}{hex_g:0>2X}{hex_b:0>2X}".format(
+                hex_r=hex_r, hex_g=hex_g, hex_b=hex_b
+            )
+
+        # color the rules - sorting by name first gives deterministic output
+        rules = sorted(self.rules, key=lambda r: r.name)
+        huefactor = 2 / (3 * len(rules))
+        rulecolor = {
+            rule.name: hsv_to_htmlhexrgb(i * huefactor, 0.6, 0.85)
+            for i, rule in enumerate(rules)
+        }
+        nodes_headers = [
+            f"\tid{index}[{node2label(node)}]" for index, node in enumerate(graph)
+        ]
+        nodes_styles = [
+            f"\tstyle id{index} fill:{rulecolor[str(node)]},stroke-width:2px,color:#333333{node2style(node)}"
+            for index, node in enumerate(graph)
+        ]
+        edges = [
+            f"\tid{index} --> id{index_dep}"
+            for index, (_, deps) in enumerate(graph.items())
+            for index_dep, _ in enumerate(deps)
+        ]
+        return (
+            textwrap.dedent(
+                """\
+            ---
+            title: DAG
+            ---
+            flowchart TB
+            """
+            )
+            + "{}\n{}\n{}".format(
+                "\n".join(nodes_headers), "\n".join(nodes_styles), "\n".join(edges)
+            )
+        )
+
     def _dot(
         self,
         graph,
@@ -3021,7 +3089,10 @@ def norm_rule_relpath(f, rule):
         return files
 
     def __str__(self):
-        return self.dot()
+        if self.workflow.dag_settings.print_dag_as == str(PrintDag.DOT):
+            return self.dot()
+        if self.workflow.dag_settings.print_dag_as == str(PrintDag.MERMAID_JS):
+            return self.mermaid_js()
 
     def __len__(self):
         return self._len
diff --git a/src/snakemake/settings/enums.py b/src/snakemake/settings/enums.py
@@ -33,3 +33,8 @@ class StrictDagEvaluation(SettingsEnumBase):
     FUNCTIONS = 0
     CYCLIC_GRAPH = 1
     PERIODIC_WILDCARDS = 2
+
+
+class PrintDag(SettingsEnumBase):
+    DOT = 0
+    MERMAID_JS = 1
diff --git a/src/snakemake/settings/types.py b/src/snakemake/settings/types.py
@@ -39,6 +39,7 @@
     CondaCleanupPkgs,
     Quietness,
     StrictDagEvaluation,
+    PrintDag,
 )
 
 
@@ -209,6 +210,7 @@ class DAGSettings(SettingsBase):
     max_inventory_wait_time: int = 20
     max_checksum_file_size: int = 1000000
     strict_evaluation: AnySet[StrictDagEvaluation] = frozenset()
+    print_dag_as: PrintDag = PrintDag.DOT
     # strict_functions_evaluation: bool = False
     # strict_cycle_evaluation: bool = False
     # strict_wildcards_recursion_evaluation: bool = False
diff --git a/src/snakemake/workflow.py b/src/snakemake/workflow.py
@@ -1263,6 +1263,8 @@ def execute(
                 and self.exec_mode == ExecMode.DEFAULT
                 and self.remote_execution_settings.job_deploy_sources
                 and not executor_plugin.common_settings.can_transfer_local_files
+                and not self.dryrun
+                and len(self.dag)
             )
             if should_deploy_sources:
                 # no shared FS, hence we have to upload the sources to the storage

Original file line number	Diff line number	Diff line change
`@@ -1263,6 +1263,8 @@ def execute(`
`1263`	`1263`	`and self.exec_mode == ExecMode.DEFAULT`
`1264`	`1264`	`and self.remote_execution_settings.job_deploy_sources`
`1265`	`1265`	`and not executor_plugin.common_settings.can_transfer_local_files`
	`1266`	`+ and not self.dryrun`
	`1267`	`+ and len(self.dag)`
`1266`	`1268`	`)`
`1267`	`1269`	`if should_deploy_sources:`
`1268`	`1270`	`# no shared FS, hence we have to upload the sources to the storage`