Standardize regex syntax

Use leading hyphen in regex character sets to follow the convention of
literal hyphens in either first or last position of a set and to support
easier appending of additional characters to the end of the set without
accidentally implementing a range with a trailing hyphen. Also, use a
compressed syntax for dates.

Snakefile

@@ -133,9 +133,9 @@ config["conda_environment"] = CONDA_ENV_PATH
 wildcard_constraints:
     # Allow build names to contain alphanumeric characters, underscores, and hyphens
     # but not special strings used for Nextstrain builds.
-    build_name = r'(?:[a-zA-Z0-9_-](?!(tip-frequencies)))+',
-    date = r"[0-9][0-9][0-9][0-9]-[0-9][0-9]-[0-9][0-9]",
-    origin = r"[a-zA-Z0-9_-]+"
+    build_name = r'(?:[-a-zA-Z0-9_](?!(tip-frequencies)))+',
+    date = r"\d{4}-\d{2}-\d{2}",
+    origin = r"[-a-zA-Z0-9_]+"
 
 localrules: clean
 

workflow/schemas/config.schema.yaml

@@ -42,7 +42,7 @@ properties:
       # Allow build names to contain alphanumeric characters, underscores, and hyphens
       # but not special strings used for Nextstrain builds.  Also used in the
       # workflow's wildcard_constraints.
-      pattern: "^(?:[a-zA-Z0-9_-](?!(tip-frequencies)))+$"
+      pattern: "^(?:[-a-zA-Z0-9_](?!(tip-frequencies)))+$"
 
   S3_DST_COMPRESSION:
     type: string

workflow/snakemake_rules/export_for_nextstrain.smk

@@ -140,8 +140,8 @@ rule dated_json:
     wildcard_constraints:
         # Allow build names to contain alphanumeric characters, underscores, and hyphens
         # but not special strings used for Nextstrain builds.
-        build_name = r'(?:[a-zA-Z0-9-_](?!(tip-frequencies|\d{4}-\d{2}-\d{2})))+',
-        date = r"[0-9][0-9][0-9][0-9]-[0-9][0-9]-[0-9][0-9]"
+        build_name = r'(?:[-a-zA-Z0-9_](?!(tip-frequencies|\d{4}-\d{2}-\d{2})))+',
+        date = r"\d{4}-\d{2}-\d{2}"
     conda: config["conda_environment"]
     shell:
         """