fix: fix false reruns after checkpoints (#1907)

fixes issue #1818

### Description

<!--Add a description of your PR here-->

### QC
<!-- Make sure that you can tick the boxes below. -->

* [x] The PR contains a test case for the changes or the changes are
already covered by an existing test case.
* [x] The documentation (`docs/`) is updated to reflect the changes or
this is not necessary (e.g. if the change does neither modify the
language nor the behavior or functionalities of Snakemake).

snakemake/dag.py

@@ -1108,11 +1108,23 @@ def update_needrun(job):
                     reason.updated_input.update(updated_input)
                 if not updated_input:
                     # check for other changes like parameters, set of input files, or code
+                    depends_on_checkpoint_target = any(
+                        f.flags.get("checkpoint_target") for f in job.input
+                    )
+
                     if "params" in self.workflow.rerun_triggers:
                         reason.params_changed = any(
                             self.workflow.persistence.params_changed(job)
                         )
-                    if "input" in self.workflow.rerun_triggers:
+                    if (
+                        "input" in self.workflow.rerun_triggers
+                        and not depends_on_checkpoint_target
+                    ):
+                        # When the job depends on a checkpoint, it will be revaluated in a second pass
+                        # after the checkpoint output has been determined.
+                        # The first pass (with depends_on_checkpoint_target == True) is not informative
+                        # for determining the input file set, as it will change after evaluating the
+                        # input function of the job in the second pass.
                         reason.input_changed = any(
                             self.workflow.persistence.input_changed(job)
                         )

tests/test_github_issue1818/.snakemake/metadata/YWdncmVnYXRlZC50eHQ=

@@ -0,0 +1 @@
+{"version": null, "code": "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", "rule": "aggregate", "input": ["my_directory/1.txt", "my_directory/2.txt", "my_directory/3.txt"], "log": [], "params": [], "shellcmd": "echo AGGREGATED > aggregated.txt", "incomplete": false, "starttime": 1665672954.609385, "endtime": 1665672954.619385, "job_hash": 8729438519112, "conda_env": null, "container_img_url": null, "input_checksums": {"my_directory/1.txt": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", "my_directory/2.txt": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", "my_directory/3.txt": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"}}

tests/test_github_issue1818/.snakemake/metadata/bXlfZGlyZWN0b3J5

@@ -0,0 +1 @@
+{"version": null, "code": "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", "rule": "somestep", "input": [], "log": [], "params": [], "shellcmd": "\n        mkdir my_directory/\n        cd my_directory\n        for i in 1 2 3; do touch $i.txt; done\n        ", "incomplete": false, "starttime": 1665672954.5860517, "endtime": 1665672954.6027184, "job_hash": 8729438582497, "conda_env": null, "container_img_url": null, "input_checksums": {}}

tests/test_github_issue1818/.snakemake/metadata/cHJvY2Vzc2VkLnR4dA==

@@ -0,0 +1 @@
+{"version": null, "code": "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", "rule": "process", "input": ["aggregated.txt"], "log": [], "params": [], "shellcmd": "echo PROCESSED > processed.txt", "incomplete": false, "starttime": 1665672954.6227183, "endtime": 1665672954.6327186, "job_hash": 8729438582463, "conda_env": null, "container_img_url": null, "input_checksums": {"aggregated.txt": "a144e9df267c10532db93a5e1625d6362d934c9973be60a6c07718a1f1942f12"}}

tests/test_github_issue1818/.snakemake/metadata/cHJvY2Vzc2VkMi50eHQ=

@@ -0,0 +1 @@
+{"version": null, "code": "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", "rule": "process2", "input": ["processed.txt"], "log": [], "params": [], "shellcmd": "echo PROCESSED2 > processed2.txt", "incomplete": false, "starttime": 1665672954.636052, "endtime": 1665672954.6493852, "job_hash": 8729438582446, "conda_env": null, "container_img_url": null, "input_checksums": {"processed.txt": "49f193a0dae7bdb2368a9f113cf2e1f3827a787dde3808a5d7838bc8ff63164c"}}

tests/test_github_issue1818/Snakefile

@@ -0,0 +1,55 @@
+# a target rule to define the desired final output
+rule all:
+    input:
+        "processed2.txt",
+
+
+# the checkpoint that shall trigger re-evaluation of the DAG
+# an number of file is created in a defined directory
+checkpoint somestep:
+    output:
+        directory("my_directory/"),
+    shell:
+        """
+        mkdir my_directory/
+        cd my_directory
+        for i in 1 2 3; do touch $i.txt; done
+        """
+
+
+# input function for rule aggregate, return paths to all files produced by the checkpoint 'somestep'
+def aggregate_input(wildcards):
+    checkpoint_output = checkpoints.somestep.get(**wildcards).output[0]
+    return expand(
+        "my_directory/{i}.txt",
+        i=glob_wildcards(os.path.join(checkpoint_output, "{i}.txt")).i,
+    )
+
+
+rule aggregate:
+    input:
+        aggregate_input,
+    output:
+        "aggregated.txt",
+    shell:
+        "echo AGGREGATED > {output}"
+
+
+# Fail here if the job runs again, as we want to ensure that snakemake does not false trigger a rerun
+# as reported in issue #1818.
+rule process:
+    input:
+        "aggregated.txt",
+    output:
+        "processed.txt",
+    shell:
+        "exit 1; echo PROCESSED > {output}"
+
+
+rule process2:
+    input:
+        "processed.txt",
+    output:
+        "processed2.txt",
+    shell:
+        "echo PROCESSED2 > {output}"

tests/test_github_issue1818/aggregated.txt

@@ -0,0 +1 @@
+AGGREGATED

tests/test_github_issue1818/expected-results/aggregated.txt

@@ -0,0 +1 @@
+AGGREGATED

tests/test_github_issue1818/expected-results/processed.txt

@@ -0,0 +1 @@
+PROCESSED

tests/test_github_issue1818/expected-results/processed2.txt

@@ -0,0 +1 @@
+PROCESSED2

tests/test_github_issue1818/processed.txt

@@ -0,0 +1 @@
+PROCESSED

tests/test_github_issue1818/processed2.txt

@@ -0,0 +1 @@
+PROCESSED2

tests/tests.py

@@ -1982,3 +1982,8 @@ def test_github_issue1618():
 
 def test_conda_python_script():
     run(dpath("test_conda_python_script"), use_conda=True)
+
+
+@skip_on_windows
+def test_github_issue1818():
+    run(dpath("test_github_issue1818"), rerun_triggers="input")