Bgc "params" & "measured" addition for gdac data source

AI_POLICY.md

@@ -0,0 +1,45 @@
+# AI Usage Policy
+
+The argopy developing team has set some rules for generative AI usage. These rules are not definitive and may be updated in the future.
+
+- **All generative AI usage in any form must be disclosed.** If AI was used to generate a significant portion of your contribution (beyond simple autocomplete), we ask that you **disclose it** within the code or PR description. You must state the tool you used along with a sentence saying that the work was AI-assisted.
+
+- **Pull requests and issues created by AI are forbidden.**
+  If AI isn't disclosed but a maintainer suspects its use within a PR or issue **creation**, the PR will be closed. AI assistance can be used for discussions inside PR or issues, but must always be supervized and reviewed by a human eye before submission.
+
+- **Media.**
+  Text and code are the only acceptable AI-generated content, per the
+  other rules in this policy.
+
+These rules apply to all contributors, including maintainers. 
+
+## Human Accountability
+
+**The human contributor is 100% responsible for the contribution.**
+
+If you submit a Pull Request that includes AI-generated code, documentation, or
+comments:
+
+- You must **fully understand** the code you submit, and the context in which it is included inside the global project or **Argopy**.
+- You must be able to explain the "**why**" behind the implementation during the
+  review process.
+- You are responsible for the long-term maintenance of that code.
+
+## The people 
+
+Please keep in mind that **Argopy** is developed and maintained by humans being.
+
+It is for us a fundamental aspect of the project that discussions (in issues or PR), even if happenning on github, are made between humans. Knowing this, it will be considered impolite to approach the **Argopy** dev community with AI-agent.
+
+## Ethics, Ecology & Expertise
+
+For more than 20 years, the Argo program has been a key element in the observation and understanding of climate change and its impact on the ocean. Argo is incremental in monitoring how the human-driven climate change negatively impact the ocean state and marine life ([check the IPCC special report on ocean](https://www.ipcc.ch/srocc/) or [that brief overview also from the IPCC](https://www.ipcc.ch/report/ar6/wg1/downloads/factsheets/IPCC_AR6_WGI_Sectoral_Fact_Sheet_Marine_Ecosystems_Fisheries.pdf)). So, it is of primary and inherent concern to Argo to limit its environmental footprint.
+
+We like to think that **Argopy** is part of the Argo ecosystem.
+
+As sustainable software developers, the **Argopy** team aims to better understand the generative AI’s environmental impact, following our preliminary work in [monitoring the carbon footprint of maintaining and developing **Argopy**](https://argopy.readthedocs.io/en/latest/energy.html). But very little information are available on the generative AI’s environmental impact and "current GAI models are deployed mainly in carbon-intensive regions" ([Ding et al, 2025](https://doi.org/10.1016/j.xinn.2025.100866)).
+
+With that, 2 things to keep in mind when contributing to the development of **Argopy** :
+
+- **Ethics & Ecology** : Not knowing the - presumably very large - impact of generative AI on CO2 emissions (and use of limited Earth resources), it is important to us that we adopt a conservative and limited approach regarding GAI usage in contributing to **Argopy** in order to limit its environmental footprint.
+- **Expertise** : **Argopy**, as part of a public research infrastructure, has more duty in developing/preserving a technical and scientific expertise for the public good, than participating in a blind race to productivity gain (in the [social acceleration sense](https://en.wikipedia.org/wiki/Social_acceleration), an argument echoing the [slow science](https://en.wikipedia.org/wiki/Slow_science) approach at work within **Argopy**). It is thus also important to us that we adopt a conservative and limited approach regarding GAI usage in contributing to **Argopy** to preserve a good level of human expertise regarding the produced code and our associated ability to disseminate our expertise to the end-users.

argopy/data_fetchers/erddap_data.py

@@ -29,7 +29,6 @@
 from .proto import ArgoDataFetcherProto
 from .erddap_data_processors import pre_process
 
-
 log = logging.getLogger("argopy.erddap.data")
 
 access_points = ["wmo", "box"]
@@ -49,6 +48,7 @@ class ErddapArgoDataFetcher(ArgoDataFetcherProto):
     This class is a prototype not meant to be instantiated directly
 
     """
+
     data_source = "erddap"
 
     ###
@@ -147,8 +147,8 @@ def __init__(  # noqa: C901
         self.fs = kwargs["fs"] if "fs" in kwargs else httpstore(**self.store_opts)
 
         self.parallelize, self.parallel_method = PARALLEL_SETUP(parallel)
-        if self.parallelize and self.parallel_method == 'thread':
-            self.parallel_method = 'erddap'  # Use our custom filestore
+        if self.parallelize and self.parallel_method == "thread":
+            self.parallel_method = "erddap"  # Use our custom filestore
         self.progress = progress
         self.chunks = chunks
         self.chunks_maxsize = chunks_maxsize
@@ -239,6 +239,17 @@ def __init__(  # noqa: C901
                         % (v, ", ".join(self._bgc_vlist_avail))
                     )
 
+            if (
+                self._bgc_vlist_params is not None
+                and self._bgc_vlist_measured is not None
+            ):
+                for v in self._bgc_vlist_measured:
+                    if v not in self._bgc_vlist_params:
+                        raise ValueError(
+                            "'%s' can't be in the 'measured' list if it is not in the 'params' list. Please check your arguments."
+                            % v
+                        )
+
     def __repr__(self):
         summary = ["<datafetcher.erddap>"]
         summary.append(self._repr_data_source)
@@ -359,7 +370,7 @@ def _minimal_vlist(self):
             ]
             [vlist.append(p) for p in plist]
 
-            # Core/Deep variables:
+            # BGC/Core/Deep variables:
             plist = [p.lower() for p in list_core_parameters()]
             [vlist.append(p) for p in plist]
             [vlist.append(p + "_qc") for p in plist]
@@ -521,7 +532,9 @@ def getNfromncHeader(url):
                 if "Your query produced no matching results. (nRows = 0)" in ncHeader:
                     return 0
                 else:
-                    lines = [line for line in ncHeader.splitlines() if "row = " in line][0]
+                    lines = [
+                        line for line in ncHeader.splitlines() if "row = " in line
+                    ][0]
                     return int(lines.split("=")[1].split(";")[0])
             except Exception:
                 raise ErddapServerError(
@@ -944,6 +957,6 @@ def uri(self):
                 except DataNotFound:
                     log.debug("This box fetcher will contain no data")
                 except ValueError as e:
-                    if 'not available for this access point' in str(e):
+                    if "not available for this access point" in str(e):
                         log.debug("This box fetcher does not contained required data")
             return urls