install from source about gcam-core

[pkyle]

gcamdata doesn't install R packages on your computer. There is a list of packages that should be installed, using the install.packages() function in R, in the file input/gcamdata/DESCRIPTION. The file isn't automatically updated, and as such the version numbers indicated in that file can be ignored. Still, every package listed in the "Imports" section should be installed locally. Copying here in case it's helpful.

assertthat
dplyr
methods
magrittr
tibble
tidyr
readr
data.table
rlang
stringr

[fipoucat]

Thank you Page, I installed the packages but was not very lucky because at one point in the install process I have another error: [ 1] "module_energy_L202.Ccoef" [1] "- make 0.01" [1] "module_energy_L210.resources" Error in `mutate()` at gcamdata/R/utils-data.R:32:3: ℹ In argument: `environCost = if_else(resource == "coal" & region %in% L210.low_reg, 0, environCost)`. Caused by error in `if_else()`: ! `false` must be a vector, not `NULL`. Backtrace: ▆ 1. ├─gcamdata::driver(write_output = FALSE, write_xml = TRUE) 2. │ └─gcamdata:::run_chunk(chunk, all_data[input_names]) at gcamdata/R/driver.R:364:7 3. │ ├─base::do.call(chunk, list(driver.MAKE, all_data)) at gcamdata/R/driver.R:16:3 4. │ └─gcamdata:::module_energy_L210.resources("MAKE", `<named list>`) 5. │ └─... %>% ... at gcamdata/R/zenergy_L210.resources.R:537:5 6. ├─gcamdata:::add_precursors(...) 7. ├─gcamdata:::add_legacy_name(., "L210.RsrcEnvironCost_SSP4", overwrite = TRUE) at gcamdata/R/utils-data.R:108:3 8. ├─gcamdata:::add_comments(., "EnvironCost adjusted for high growth and low growth regions ") at gcamdata/R/utils-data.R:67:3 9. ├─gcamdata:::add_comments(., "A10.EnvironCost_SSPs written to all regions") at gcamdata/R/utils-data.R:49:3 10. ├─gcamdata:::add_units(., "$/GJ") at gcamdata/R/utils-data.R:49:3 11. ├─gcamdata:::add_title(...) at gcamdata/R/utils-data.R:91:3 12. ├─dplyr::mutate(...) at gcamdata/R/utils-data.R:32:3 13. ├─dplyr:::mutate.data.frame(...) 14. │ └─dplyr:::mutate_cols(.data, dplyr_quosures(...), by) 15. │ ├─base::withCallingHandlers(...) 16. │ └─dplyr:::mutate_col(dots[[i]], data, mask, new_columns) 17. │ └─mask$eval_all_mutate(quo) 18. │ └─dplyr (local) eval() 19. ├─dplyr::if_else(resource == "coal" & region %in% L210.low_reg, 0, environCost) 20. │ └─dplyr:::vec_case_when(...) 21. │ └─vctrs::list_check_all_vectors(values, arg = values_arg, call = call) 22. └─vctrs:::stop_scalar_type(`<fn>`(NULL), "false", `<env>`) 23. └─vctrs:::stop_vctrs(...) 24. └─rlang::abort(message, class = c(class, "vctrs_error"), ..., call = call) Execution halted make: *** [Makefile:5: xml] Error 1

…

On Sun, Apr 28, 2024 at 7:27 AM Page Kyle ***@***.***> wrote: gcamdata doesn't install R packages on your computer. There is a list of packages that should be installed, using the install.packages() function in R, in the file input/gcamdata/DESCRIPTION. The file isn't automatically updated, and as such the version numbers indicated in that file can be ignored. Still, every package listed in the "Imports" section should be installed locally. Copying here in case it's helpful. assertthat dplyr methods magrittr tibble tidyr readr data.table rlang stringr — Reply to this email directly, view it on GitHub <#421 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAXY6LLBCATKJCVHACXFL2LY7SQFHAVCNFSM6AAAAABG4EM656VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOBRGM3TAOBXGE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[pkyle]

Right, that's because we never followed up the GCAM7 release last summer with another release that fixes this issue. I think that code didn't crash in some older versions of dplyr, for some reason, because it worked when it was tested out! This exact error is shown here: #393 (comment)
And my first reply is what I'd recommend doing here too: #393 (comment)

[fipoucat]

There is an issue cloning the repository. git remote add github_pkyle https://github.com/pkyle/gcam-core.git fatal: not a git repository (or any parent up to mount point /media/sarr) Stopping at filesystem boundary (GIT_DISCOVERY_ACROSS_FILESYSTEM not set).

…

On Mon, Apr 29, 2024 at 3:50 PM Page Kyle ***@***.***> wrote: Right, that's because we never followed up the GCAM7 release last summer with another release that fixes this issue. I think that code didn't crash in some older versions of dplyr, for some reason, because it worked when it was tested out! This exact error is shown here: #393 (comment) <#393 (comment)> And my first reply is what I'd recommend doing here too: #393 (comment) <#393 (comment)> — Reply to this email directly, view it on GitHub <#421 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAXY6LM6JFZTFVFDHZR7YJLY7ZT5DAVCNFSM6AAAAABG4EM656VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOBTGA4DQOBRGM> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[pkyle]

Oh, that git remote ... command can only be executed from within a git repo. So, your working directory should be gcam-core (or one of the subdirectories). If you're working from a static release version, as opposed to a locally cloned git repository, I'm not sure that there's any simple/clean way to automatically apply (cherry-pick) these commits, but they're pretty small, and can just be done manually. Here are the two that you'll want:
pkyle@f3377fc
pkyle@1d25fbc

[fipoucat]

Well, leave it there for now, waiting another release

…

On Wed, May 1, 2024 at 7:16 PM Page Kyle ***@***.***> wrote: Oh, that git remote ... command can only be executed from within a git repo. So, your working directory should be gcam-core (or one of the subdirectories). If you're working from a static release version, as opposed to a locally cloned git repository, I'm not sure that there's any simple/clean way to automatically apply (cherry-pick) these commits, but they're pretty small, and can just be done manually. Here are the two that you'll want: ***@***.*** <pkyle@f3377fc> ***@***.*** <pkyle@1d25fbc> — Reply to this email directly, view it on GitHub <#421 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAXY6LPL2O6SCWDLVCUCLJDZAE5SNAVCNFSM6AAAAABG4EM656VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOBYHE2TOMJSGU> . You are receiving this because you authored the thread.Message ID: ***@***.***>