poncho_package_create(1)

🦋 poncho_package_create — the NAME section, bestie

okay so this little cutie lives in the Cooperative Computing Tools manual, and its whole job is:

poncho_package_create — a command-line utility for creating a Conda virtual environment given a Python dependencies file 💅

that's it. that's the name. she's a package-maker. she makes environments. periodt. ✨

---

✨ SYNOPSIS ✨

first, the vibe: here's what a specification file can look like, okay? like, this is the shape of the JSON you feed her:

Example Specification:
{
    "conda": {
        "channels": [
            "conda-forge"
        ],
        "dependencies": [
            "matplotlib=3.7.1=py311h38be061_0",
            "numpy=1.24.2=py311h8e6699e_0",
            "pip=23.0.1=pyhd8ed1ab_0",
            "python=3.11.0=he550d4f_1_cpython",
            {
                "pip": [
                    "uproot==5.0.5"
                ]
            }
        ]
    },

    "git": {
                "DATA_DIR": {
                        "remote": "http://.../repo.git",
                }
    },

    "http": {
                "REFERENCE_DB": {
                        "type": "file",
                        "url": "https://.../example.dat"
                },
                "TRAINING_DATASET": {
                        "type": "tar",
                        "compression": "gzip",
                        "url": "http://.../dataset.tar.gz"
                }
    }
}

and then the actual command line? literally:

poncho_package_create [options] <dependency-file> <<output-path>>

yes. both sets of angle brackets. that's canon, don't @ me 🫶

---

🌸 DESCRIPTION 🌸

so poncho_package_create is a simple command-line utility that creates a local Conda environment from an input JSON dependency file. no cap, that's the whole premise.

the command creates an environment tarball at output-path that can be sent to and run on different machines — as long as those machines have the same architecture. (same arch or it's not gonna slay, sorry bestie 😭)

the dependency-file argument is the path (relative or absolute) to the JSON specification file. yes, either kind of path works, she's flexible like that.

the output-path argument specifies the path for the environment tarball that gets created — and it should usually end in .tar.gz. like, strongly usually. it's giving "please name it right" 💅

---

💅 OPTIONS 💅

• --conda-executable=<path> — where's your conda executable at?? point her to it with this. if you don't give her one, she'll try mamba, then $CONDA_EXE, then conda, then microconda — in that exact order. it's a whole audition lineup 🎤
• --no-microconda — do NOT go downloading microconda if a conda executable can't be found. like, hands off the downloads, tyvm 🙈
• -h, --help — show the help message. she'll explain herself. iconic behavior 👑

---

🔥 EXIT STATUS 🔥

• success → returns zero ✨
• failure → returns non-zero 💀

that's the whole status report, she keeps it simple.

---

🍓 EXAMPLE 🍓

poncho_package_create dependencies.json example_venv.tar.gz

This will create an example_venv.tar.gz environment tarball within the current working directory, which can then be exported to different machines for execution. so like — build once, yeet it everywhere, run it there. main character energy fr fr 🌈

---

👑 COPYRIGHT 👑

The Cooperative Computing Tools are Copyright (C) 2022 The University of Notre Dame. This software is distributed under the GNU General Public License. See the file COPYING for details.

---

okay that's the whole page bestie, you ate it, go make some environments 🧚

CCTools 7.15.9 FINAL — poncho_package_create(1) — mwah, that's a wrap, ily 💜