Update sheet2linkml so that it can be run from the root directory of this package

.gitignore

@@ -10,6 +10,7 @@ local
 # Google API authentication information
 google_api_credentials.json
 token.pickle
+sheets.googleapis.com-python.json
 
 # From LinkML template (https://github.com/linkml/linkml-template/blob/1c94efc880b607b36786b87e164455a099618ffa/.gitignore)
 Pipfile

Makefile

@@ -44,6 +44,7 @@ PKG_T_CSV = $(PKG_DIR)/csv
 ENV = PIPENV_IGNORE_INSTALLED=1
 RUN = $(ENV) pipenv run
 GEN_OPTS = --log_level WARNING
+CDM_GOOGLE_SHEET_ID=1oWS7cao-fgz2MKWtyr8h2dEL9unX__0bJrWKv6mQmM4
 
 # ----------------------------------------
 # TOP LEVEL TARGETS
@@ -304,17 +305,16 @@ docserve: gen-docs
 
 # Regenerate from Google Sheets. Note that this uses a *separate* Pipenv in the
 # generators/google-sheets directory, so we have to run pipenv install on it separately.
-regen-google-sheets:
-	cd generators/google-sheets && pipenv install && pipenv run python sheet2linkml.py && cp output/CDM_Dictionary_v1_Active.yaml ../../model/schema/crdch_model.yaml && cd -
+generate-model: install
+	CDM_GOOGLE_SHEET_ID=$(CDM_GOOGLE_SHEET_ID) $(RUN) python vendor/sheet2linkml/sheet2linkml.py --output model/schema/crdch_model.yaml
 
 # MAM 20210806 not sure how this fits into the linkml model template's doc building/publsihing approach
-# shoudn't it be using $(RUN) not ${RUN} ?
 # Deploy changes to the `dev` version on the gh-pages branch.
 # Note that this is not dependent on stage-docs, since you
 # would generate the docs (usually as part of a `make all`)
 # before you deploy it in a separate step.
-#gh-deploy: pipenv-install
-#	${RUN} mike deploy dev -p
+gh-deploy: install
+	$(RUN) mike deploy dev -p
 
 #### MAM 20210729
 pypi:

Pipfile

@@ -4,24 +4,35 @@ verify_ssl = true
 name = "pypi"
 
 [packages]
-linkml = {editable = true, ref = "ccdh-dev", git = "https://github.com/cancerDHC/linkml.git"} # required by cancerDHC
-linkml-runtime = "~=1.0, >=1.0.3"
+# For LinkML, we use the `ccdh-dev` branch, which we manually keep in sync with LinkML's main branch
+# as well as some additional changes required for CRDCH generation.
+linkml = {editable = true, ref = "ccdh-dev", git = "https://github.com/cancerDHC/linkml.git"}
+
+# Since we use the development version of LinkML, we also use the development version of linkml-runtime.
+# This may sometimes cause issues between this and LinkML.
+linkml-runtime = {editable = true, git = "https://github.com/linkml/linkml-runtime.git"}
+
 pyparsing = "<=2.4.7"
 rdflib-pyld-compat = "*" # resinstating after removal by migration to linkml model template
 
 [dev-packages]
-black = "==v21.5b1" # from old generators/google-sheets/Pipfile
-mike = "*"
-mkdocs-material = "*"
-pydantic = {extras = ["dotenv"], version = "*"} # from old generators/google-sheets/Pipfile
-pygsheets = "*" # from old generators/google-sheets/Pipfile
-pyparsing = "<=2.4.7" # indirectly from old generators/google-sheets/Pipfile
+pyyaml = "*"            # For reading and writing YAML files.
+
+# Packages needed for testing.
 pytest = "~=6.2.4"
-pyyaml = "*" # from old generators/google-sheets/Pipfile
-requests = "*" # from old generators/google-sheets/Pipfile
 tox = "*"
 twine = "*" # should this go in make-venv/Pipfile ?
 
+# Packages needed by sheet2linkml.
+black = "==v21.5b1"     # Used for code linting.
+requests = "*"          # For making HTTP requests.
+pygsheets = "*"         # For accessing Google Sheets.
+python-dotenv = "*"     # For reading .env files.
+
+# Packages needed for building and publishing documentation.
+mike = "*"
+mkdocs-material = "*"
+
 [pipenv]
 allow_prereleases = true # added by migration to linkml model template
 

README.md

@@ -10,7 +10,22 @@ Model documentation in Markdown can also be generated for this repository, and i
 
 ### Automated generation of YAML
 
-The model YAML files are programmatically generated from [the Google Sheet](https://docs.google.com/spreadsheets/d/1oWS7cao-fgz2MKWtyr8h2dEL9unX__0bJrWKv6mQmM4/). The code that generates the YAMLs are in the [`./generators/google-sheets`](./generators/google-sheets) directory. The `regen-google-sheet` target in the Makefile can update the schema from Google Sheets automatically.
+The CRDC-H model is currently in development on [a Google Sheet](https://docs.google.com/spreadsheets/d/1oWS7cao-fgz2MKWtyr8h2dEL9unX__0bJrWKv6mQmM4/).
+This repository contains a program in the [`./generators/google-sheets`](./generators/google-sheets) directory,
+called `sheet2linkml`, which can read the Google Sheet using the Google Drive API and generate a LinkML representation
+of the model.
+
+In order to run this program, you will need to create and download Google Drive client credentials. First,
+[enable the Google Drive API](https://developers.google.com/drive/api/v3/enable-drive-api). After the API is enabled, 
+[create and download the client credentials](https://www.iperiusbackup.net/en/how-to-enable-google-drive-api-and-get-client-credentials/)
+from the [Google API Console](https://console.developers.google.com/). Save the file as `google_api_credentials.json` in
+the root directory of this project. [Detailed instructions and screenshots](https://pygsheets.readthedocs.io/en/stable/authorization.html)
+are also available from the [`pygsheets` documentation](https://pygsheets.readthedocs.io/), which is the package we use to
+access Google Sheets.
+
+Once this is setup, run `make generate-model` to update the schema from Google Sheets. The first time you run this script,
+you will see a browser page asking you to log in. Follow the instructions. The script will download a token and store it
+locally. You won't need to log in in the future.
 
 ### Generation of LinkML artifacts