nfdocs-parser/nfdocs-parser.py
Thomas A. Christensen II 557a7e3ccd
Add a function that can parse recursive tuples
Signed-off-by: Thomas A. Christensen II <25492070+MillironX@users.noreply.github.com>
2022-01-05 15:19:47 -06:00

87 lines
2.7 KiB
Python
Executable file

#!/usr/bin/env python
import sys
import yaml
# Declare the docstring starting characters
DOC_STARTER = "/// "
def definition_type(signature):
# Returns "name", workflow|process|function
def_type = "unknown"
if "workflow" in signature:
def_type = "workflow"
elif "process" in signature:
def_type = "process"
elif "function" in signature:
def_type = "function"
# Check if any signature was recognized
if def_type == "unknown":
return "unknown", "an error occurred"
# Parse out the definition name
def_name = signature.replace(def_type, "").replace("{", "").strip()
# Return the results
return def_name, def_type
def params_to_list(params):
if "tuple" in params.keys():
tuple_item = nodes.list_item(text="tuple")
tuple_list = nodes.bullet_list()
for io in params["tuple"]:
tuple_list += params_to_list(io)
tuple_item += tuple_list
return tuple_item
else:
return nodes.list_item(text="not tuple")
print("not yet implemented")
# Take path as single argument for now
nextflow_path = sys.argv[1]
with open(nextflow_path) as nextflow_file:
# Split by lines
nextflow_lines = nextflow_file.readlines()
# Declare some variables to keep track of where the docstrings begin and end
doc_start = 0
doc_end = 0
# Declare dictionaries to keep track of the docstrings
docstring_positions = []
# Calculate the start and end positions of each docstring
for i, line in enumerate(nextflow_lines):
# Check if this is a docstring
if line.startswith(DOC_STARTER):
# It is: check the next and previous lines to see if this is part of a block
line_previous = nextflow_lines[i-1]
line_next = nextflow_lines[i+1]
if not line_previous.startswith(DOC_STARTER):
doc_start = i
if not line_next.startswith(DOC_STARTER):
doc_end = i
# Check if we've reached the end of a docstring block
if doc_end == i:
# Add this docstring position to the array
docstring_positions.append(range(doc_start, doc_end))
# Create dictionaries for each of the block types
docstrings = {
"process": {},
"workflow": {},
"function": {}
}
# Parse out the docstrings and put them in the appropriate dictionary
for pos in docstring_positions:
proc_name, proc_type = definition_type(nextflow_lines[pos[-1]+2])
doc_yaml = ""
for i in pos:
doc_yaml = doc_yaml + nextflow_lines[i].replace(DOC_STARTER, "")
docstrings[proc_type][proc_name] = yaml.safe_load(doc_yaml)
# Display the results so far
print(docstrings)