new command geturl

doc/src/geturl.rst

@@ -0,0 +1,72 @@
+.. index:: geturl
+
+geturl command
+==============
+
+Syntax
+""""""
+
+.. code-block:: LAMMPS
+
+   geturl url keyword args ...
+
+* url = URL of the file to download
+* zero or more keyword argument pairs may be provided
+* keyword = *output* or *verify* or *overwrite*
+
+  .. parsed-literal::
+
+     *output* filename = write to *filename* instead of inferring the name from the URL
+     *verify* yes/no = verify SSL certificate and hostname if *yes*, do not if *no*
+     *overwrite* yes/no = if *yes* overwrite the output file in case it exists, do not if *no*
+
+Examples
+""""""""
+
+.. code-block:: LAMMPS
+
+   geturl https://www.ctcms.nist.gov/potentials/Download/1990--Ackland-G-J-Vitek-V--Cu/2/Cu2.eam.fs
+   geturl https://github.com/lammps/lammps/blob/develop/bench/in.lj output in.bench-lj
+
+Description
+"""""""""""
+
+Download a file from an URL to the local disk. This is implemented with
+the `libcurl library <https:://curl.se/libcurl/>`_ which supports a
+large variety of protocols including "http", "https", "ftp", "scp",
+"sftp", "file".  The transfer will only be performed on MPI rank 0.
+
+The *output* keyword can be used to set the filename. By default, the last part
+of the URL is used.
+
+The *verify* keyword allows to turn on or off whether ``libcurl`` will validate
+the SSL certificate and hostname for encrypted connections.  Turning this off
+may be required when using a proxy or connecting to a server with a self-signed
+SSL certificate.
+
+The *overwrite* keyword determines whether a file should be overwritten if it
+already exists.  If the argument is *no*, then the download will be skipped
+if the file exists.
+
+----------
+
+Restrictions
+""""""""""""
+
+This command is part of the EXTRA-COMMAND package.  It is only enabled
+if LAMMPS was built with that package.  See the :doc:`Build package
+<Build_package>` page for more info.  It also requires that LAMMPS was
+built with support for `the libcurl library
+<https://curl.se/libcurl/>`_.  See the page about :ref:`Compiling LAMMPS
+with libcurl support <libcurl>` for further info.  If support for
+libcurl is not included, using *geturl* will trigger an error.
+
+Related commands
+""""""""""""""""
+
+:doc:`shell <shell>`
+
+Default
+"""""""
+
+*verify* = yes, *overwrite* = yes

src/EXTRA-COMMAND/geturl.cpp

@@ -0,0 +1,111 @@
+/* ----------------------------------------------------------------------
+   LAMMPS - Large-scale Atomic/Molecular Massively Parallel Simulator
+   https://www.lammps.org/, Sandia National Laboratories
+   LAMMPS development team: developers@lammps.org
+
+   Copyright (2003) Sandia Corporation.  Under the terms of Contract
+   DE-AC04-94AL85000 with Sandia Corporation, the U.S. Government retains
+   certain rights in this software.  This software is distributed under
+   the GNU General Public License.
+
+   See the README file in the top-level LAMMPS directory.
+------------------------------------------------------------------------- */
+
+/* ----------------------------------------------------------------------
+   Contributing authors:  Axel Kohlmeyer (Temple U),
+------------------------------------------------------------------------- */
+
+#include "geturl.h"
+
+#include "comm.h"
+#include "error.h"
+
+#if defined(LAMMPS_CURL)
+#include <curl/curl.h>
+#endif
+
+using namespace LAMMPS_NS;
+
+/* ---------------------------------------------------------------------- */
+
+void GetURL::command(int narg, char **arg)
+{
+#if !defined(LAMMPS_CURL)
+  error->all(FLERR, "LAMMPS has not been compiled with libcurl support");
+#else
+  if (narg < 1) utils::missing_cmd_args(FLERR, "geturl", error);
+  int verify = 1;
+  int overwrite = 1;
+
+  // process arguments
+
+  std::string url = arg[0];
+
+  // sanity check
+
+  if ((url.find(':') == std::string::npos) || (url.find('/') == std::string::npos))
+    error->all(FLERR, "URL '{}' is not a supported URL", url);
+
+  std::string output = url.substr(url.find_last_of('/') + 1);
+  if (output.empty()) error->all(FLERR, "URL '{}' must end in a file string", url);
+
+  int iarg = 1;
+  while (iarg < narg) {
+    if (strcmp(arg[iarg], "output") == 0) {
+      if (iarg + 2 > narg) utils::missing_cmd_args(FLERR, "geturl output", error);
+      output = arg[iarg + 1];
+      ++iarg;
+    } else if (strcmp(arg[iarg], "overwrite") == 0) {
+      if (iarg + 2 > narg) utils::missing_cmd_args(FLERR, "geturl overwrite", error);
+      overwrite = utils::logical(FLERR, arg[iarg + 1], false, lmp);
+      ++iarg;
+    } else if (strcmp(arg[iarg], "verify") == 0) {
+      if (iarg + 2 > narg) utils::missing_cmd_args(FLERR, "geturl verify", error);
+      verify = utils::logical(FLERR, arg[iarg + 1], false, lmp);
+      ++iarg;
+    } else {
+      error->all(FLERR, "Unknown geturl keyword: {}", arg[iarg]);
+    }
+    ++iarg;
+  }
+
+  // only download files from rank 0
+
+  if (comm->me != 0) return;
+
+  if (!overwrite && platform::file_is_readable(output)) return;
+
+  // open output file for writing
+
+  FILE *out = fopen(output.c_str(), "wb");
+  if (!out)
+    error->all(FLERR, "Cannot open output file {} for writing: {}", output, utils::getsyserror());
+
+  // initialize curl and perform download
+
+  CURL *curl;
+  curl_global_init(CURL_GLOBAL_DEFAULT);
+  curl = curl_easy_init();
+  if (curl) {
+    curl_easy_setopt(curl, CURLOPT_URL, url.c_str());
+    curl_easy_setopt(curl, CURLOPT_WRITEDATA, (void *) out);
+    curl_easy_setopt(curl, CURLOPT_FILETIME, 1L);
+    curl_easy_setopt(curl, CURLOPT_FAILONERROR, 1L);
+
+    if (!verify) {
+      curl_easy_setopt(curl, CURLOPT_SSL_VERIFYPEER, 0L);
+      curl_easy_setopt(curl, CURLOPT_SSL_VERIFYHOST, 0L);
+    }
+    auto res = curl_easy_perform(curl);
+    if (res != CURLE_OK) {
+      long response = 0L;
+      curl_easy_getinfo(curl, CURLINFO_RESPONSE_CODE, &response);
+      error->one(FLERR, "Download of {} failed with: {} {}", output, curl_easy_strerror(res),
+                 response);
+    }
+    curl_easy_cleanup(curl);
+  }
+  curl_global_cleanup();
+  fclose(out);
+#endif
+}

src/EXTRA-COMMAND/geturl.h

@@ -0,0 +1,34 @@
+/* -*- c++ -*- ----------------------------------------------------------
+   LAMMPS - Large-scale Atomic/Molecular Massively Parallel Simulator
+   https://www.lammps.org/, Sandia National Laboratories
+   LAMMPS development team: developers@lammps.org
+
+   Copyright (2003) Sandia Corporation.  Under the terms of Contract
+   DE-AC04-94AL85000 with Sandia Corporation, the U.S. Government retains
+   certain rights in this software.  This software is distributed under
+   the GNU General Public License.
+
+   See the README file in the top-level LAMMPS directory.
+------------------------------------------------------------------------- */
+
+#ifdef COMMAND_CLASS
+// clang-format off
+CommandStyle(geturl,GetURL);
+// clang-format on
+#else
+
+#ifndef LMP_GETURL_H
+#define LMP_GETURL_H
+
+#include "command.h"
+
+namespace LAMMPS_NS {
+
+class GetURL : public Command {
+ public:
+  GetURL(class LAMMPS *lmp) : Command(lmp) {};
+  void command(int, char **) override;
+};
+}    // namespace LAMMPS_NS
+#endif
+#endif