From 0d40d977e52767b699c5fa3f64f1019e8bc45552 Mon Sep 17 00:00:00 2001
From: Stephen Turner <vustephen@gmail.com>
Date: Fri, 27 Sep 2024 05:15:25 -0400
Subject: [PATCH 1/2] add tibble

---
 DESCRIPTION | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/DESCRIPTION b/DESCRIPTION
index e3a54d7..139ec11 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -15,14 +15,15 @@ Imports:
     ollamar (>= 1.2.1),
     rlang,
     rmarkdown,
+    tibble,
     tidyRSS,
     tinytable
 Suggests: 
     knitr,
     markdown,
     testthat (>= 3.0.0)
+Config/testthat/edition: 3
 Encoding: UTF-8
 LazyData: true
 Roxygen: list(markdown = TRUE)
 RoxygenNote: 7.3.2
-Config/testthat/edition: 3

From d0d2be9d4a113a3bd1e5a33a1cf3805833db32ad Mon Sep 17 00:00:00 2001
From: Stephen Turner <vustephen@gmail.com>
Date: Fri, 27 Sep 2024 05:15:44 -0400
Subject: [PATCH 2/2] - Fixed list of medRxiv subjects (see `?subjects`). -
 Fixed an issue with `get_preprints()` where a valid subject has no preprints
 associated with it at all.

---
 NAMESPACE                |   1 +
 NEWS.md                  |   2 ++
 R/biorecap.R             |  48 ++++++++++++++++++++++++++++-----------
 data-raw/biorecap_data.R |   4 ++--
 data/subjects.rda        | Bin 824 -> 819 bytes
 man/safely_query_rss.Rd  |  19 ++++++++++++++++
 6 files changed, 59 insertions(+), 15 deletions(-)
 create mode 100644 man/safely_query_rss.Rd

diff --git a/NAMESPACE b/NAMESPACE
index 4fe132c..9420a28 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -8,6 +8,7 @@ export(build_prompt_preprint)
 export(build_prompt_subject)
 export(get_preprints)
 export(list_models)
+export(safely_query_rss)
 export(test_connection)
 export(tt_preprints)
 importFrom(ollamar,list_models)
diff --git a/NEWS.md b/NEWS.md
index e4b0d91..8c4200b 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -1,6 +1,8 @@
 # biorecap 0.2.1
 
 - Added a `host` argument to `biorecap_report()` and `add_summary()` that gets passed to `ollamar::generate()`, allowing the user to choose the host for ollama. By default, `host=NULL` uses ollama's default base URL (using the local ollama server). This can be used to connect to a llama server running remotely or within a firewalled network (fixes #7).
+- Fixed list of medRxiv subjects (see `?subjects`).
+- Fixed an issue with `get_preprints()` where a valid subject has no preprints associated with it at all.
 - Updated tests.
 
 # biorecap 0.2.0
diff --git a/R/biorecap.R b/R/biorecap.R
index 799a14e..9b54648 100644
--- a/R/biorecap.R
+++ b/R/biorecap.R
@@ -57,6 +57,37 @@ build_prompt_subject <- function(subject,
   return(prompt)
 }
 
+#' Safely query bioRxiv/medRxiv RSS feeds
+#'
+#' @param subject A character vector of valid bioRxiv and/or medRxiv subjects. See [subjects].
+#' @param server A character vector of either "biorxiv" or "medrxiv".
+#'
+#' @return A data frame of preprints from bioRxiv and/or medRxiv.
+#' @export
+#'
+safely_query_rss <- function(subject, server=c("biorxiv", "medrxiv")) {
+  server <- rlang::arg_match(server)
+  baseurl <- sprintf("https://connect.%s.org/%s_xml.php?subject=", server, server)
+  out <-
+    lapply(subject, \(x) {
+      tryCatch({
+        suppressMessages(preprints <- tidyRSS::tidyfeed(paste0(baseurl, x)))
+      },
+      error = function(e) {
+        message("No preprints found for subject: ", x)
+        tibble::tibble(feed_title=character(), feed_link=character(), feed_description=character(), item_title=character(),
+                       item_link=character(), item_description=character(), item_category=list())
+      })
+    })
+  out <-
+    out |>
+    stats::setNames(subject) |>
+    dplyr::bind_rows(.id="subject") |>
+    dplyr::select("subject", title="item_title", url="item_link", abstract="item_description") |>
+    dplyr::mutate(dplyr::across(dplyr::everything(), trimws)) |>
+    dplyr::mutate("source"=server, .before=1)
+}
+
 #' Get bioRxiv/medRxiv preprints
 #'
 #' @param subject A character vector of valid bioRxiv and/or medRxiv subjects. See [subjects].
@@ -80,26 +111,17 @@ get_preprints <- function(subject="all", clean=TRUE) {
   subject_bio <- subject[subject %in% biorecap::subjects$biorxiv]
   if (length(subject_bio)>0) {
     preprints$bio <-
-      lapply(subject_bio, \(x) suppressMessages(preprints <- tidyRSS::tidyfeed(paste0("https://connect.biorxiv.org/biorxiv_xml.php?subject=", x)))) |>
-      stats::setNames(subject_bio) |>
-      dplyr::bind_rows(.id="subject") |>
-      dplyr::select("subject", title="item_title", url="item_link", abstract="item_description") |>
-      dplyr::mutate(dplyr::across(dplyr::everything(), trimws)) |>
+      safely_query_rss(subject=subject_bio, server="biorxiv") |>
       dplyr::mutate("source"="bioRxiv", .before=1)
-    if (nrow(preprints$bio)<1L) stop("Something went wrong. No papers found for subject ", subject) #nocov
+    if (nrow(preprints$bio)<1L) warning("Something went wrong. No papers found for subject ", subject) #nocov
   }
 
-
   subject_med <- subject[subject %in% biorecap::subjects$medrxiv]
   if (length(subject_med)>0) {
     preprints$med <-
-      lapply(subject_med, \(x) suppressMessages(preprints <- tidyRSS::tidyfeed(paste0("https://connect.medrxiv.org/medrxiv_xml.php?subject=", x)))) |>
-      stats::setNames(subject_med) |>
-      dplyr::bind_rows(.id="subject") |>
-      dplyr::select("subject", title="item_title", url="item_link", abstract="item_description") |>
-      dplyr::mutate(dplyr::across(dplyr::everything(), trimws)) |>
+      safely_query_rss(subject=subject_med, server="medrxiv") |>
       dplyr::mutate("source"="medRxiv", .before=1)
-    if (nrow(preprints$med)<1L) stop("Something went wrong. No papers found for subject ", subject) #nocov
+    if (nrow(preprints$med)<1L) warning("Something went wrong. No papers found for subject ", subject) #nocov
   }
 
   preprints <- dplyr::bind_rows(preprints)
diff --git a/data-raw/biorecap_data.R b/data-raw/biorecap_data.R
index d6d673b..2ee228d 100644
--- a/data-raw/biorecap_data.R
+++ b/data-raw/biorecap_data.R
@@ -39,10 +39,10 @@ subjects$medrxiv <- c("all",
                       "Cardiovascular_Medicine",
                       "Dentistry_and_Oral_Medicine",
                       "Dermatology",
-                      "Dermatology",
+                      "Emergency_Medicine",
                       "endocrinology",
                       "Epidemiology",
-                      "ecology",
+                      "endocrinology",
                       "epidemiology",
                       "Forensic_Medicine",
                       "Gastroenterology",
diff --git a/data/subjects.rda b/data/subjects.rda
index a1dd9d55a2892e9cef02b2b0242adc496eace161..c5c6603a1bd53fa9de8df62c7283e9c48c3f46c2 100644
GIT binary patch
delta 809
zcmV+^1J?Yw2D1hZLRx4!F+o`-Q(1Gw3I>r58Gk<Mb~|9_Ez~qk2|XiC4FEkLXaE2J
z15FYcG}SR4r|ONQ(9jtGU?->m)buoyRMSB98U}%&00xGJKyv_?0WbgwfB={P2@(iQ
zhK&RX=z~odjW(cYG!SdX`^y-Cek=Tv5mpKL|Fa^-Re%ciyEscp3y+p&7Ry)5EEa+o
zIe)1XY8N0Tg7qN+OnF48I3T3B&a#5FinSJKFlbv)gft`#jA(|E5e)(tkAw*#Yy~Id
zmj4_4^y81z#<Wci2xgG#`E~9Qt?9uuGFVI`ph$CJV(|KuQp&{dYsPo3KdS#7f=F+B
zB~<1fjbyN*So^>OZ_^E98D&D^+-a)3Gk=sd+?=#cUkd*{Jzd`dNlokR+)m^w@v%^y
zHeB$RInY%K-v_qYk))hjM9`&sE!jIz7Q*cK@wEZfT{v7)=4X<<psVIkZc9P6koxxf
zmk5Z>fw16;D31(;l8G;H1}d1I$|SbVqe?fEqmZtfCtR9|!R9s=H7brMYkRP3RDT}P
z-Lgh*r<#ee0Rn86D{BppQ4?5Z-}?65a9<I@!!Hb}%J*-q-wM~3b{SESh5ac(-4&$e
zVJovOxE0~EJ<*G@#4eAwB{bVi<Q{kdiyLnpA|t89wN5EJZPeo<EcGv;tQ$dr)@mg(
zlE~?z=xZto1f&FLD492RPrq^(HGgzv7vJh+f#7DwCT&y{n(}kDA^;gQF0+mxX269W
zSZmGpcm_j|i6FyC5v^y#f7Xj*5XWvsC<E@_iw#m{1+j0{II&>YmIR41D)9ir;3Xg-
zjj2oXrh+rOh_uvPbT1%M2VI)be70rVuCnSHifRUnR<=LmJ<n9;pn)l@NPof~SpeGF
zXI08Fu&0tj93m2^#<blUjD=Q~4qfa3Tnr%W3#RVtR|9^&PNBji_{{H}mYsdM9sHTE
zMV^8OwJr4}JTNrDYmu^%G6|UuR~EFfNM@3-gW?#%G4uYT70FPp<XSCo!-_JN3T;j_
nPEs@x+>s3-w6nBc<i^Hxxe{PGmy8iB$N0OFDZ+$<o*+;#2&Hdr

delta 814
zcmV+}1JV4m2Dk<eLRx4!F+o`-Q(5D8+J=!18Gk*}%iiaE)^x4IK!Hu9%6c?48K@Z=
zWY8K0fDn;ACZ4ACCTa!~Mg-7kG8r_}Ks3-zHBA)TRQ#rqnIWJI38tC`)b#>HiKv?<
znwwKV0009(Xfy&+OiAe)Xf$X8KmY&$1Q~9wvja#M#XlqnqXhK-Vnzm0fC{#{H|3%d
z=YPBMqP!)%ETT7%p@@<ZL1F|emXaV6XQEXOWED%wbd(gNQK-RbV-n1SA%Z~I#)xSl
z5X2#8@Ps5yfTZ)^v7Ws-F1ysai3;opRj_Ev%^o5Q<qQc!VxrFx4`9ZH8;HRJ)fCU&
zcnxt&zOa5ZRFUwClpoGJmUPxdIPd@qcYn^nV8k8-gll&b@1!_cA~nLmP=!a`T-$dA
zB4e{=!1RGF-zfp(IyzgJZ59D9+IYp>FA#OQx&&+p>Wpv$o=m$A4gkFU2w-dqC#@$a
z2i0XBBt(d+2S>Z$eee-6jte3pA~Hot0X1w4rUV>^0?U?0bq*#-9Xn(xT5Xh0iGP}$
z5e1Up7OQ#?@b>hX>%#o*$fiJmV+&bYU}HWYj8nmv=c-ZygiFl~PzEPlc{06@_rB(l
zln|gN0T2x>q1r=MgyYgJ&z`DXW@H50<LHhn)r^fcE1FHWc0m!s;##wcr-u%!i_7uQ
zuA?Y6f)g1?TL!K&R;9HhZA3s?z<+unsG*7=mpp)*bhi5I<7GhXF<lcgDupJrIpKx@
z69v@wXER0+5v-#dWlZM5h=NEiwjuS?YRmV&wz{56A|MUv%Vw#83ZSmeWte&eU9v0*
zO9_W?M(IUhgHzVO>1l;WpQN_f)lZc|D3*OU4HxFWzjdu)p=p_`Cg*Pc#((@4_F*lH
zjG+ezk=u;tN>h<{2GA%tc!?WhOF@y)Vp4+^Y5+#%4t(PFzg12P!MZ*}omls{>UczI
z*@$-L!9fUU2oBOLuL{$csRT_%HzOD`$YGJoW>|q<%}T<rfMZx;*Zz!oQbvwcSk1<~
sS0>VqQgLkMBGNknLj>5dv=>(7#>Q?t2}EWYkL3O?<ce^iA;#~u46HF|*Z=?k

diff --git a/man/safely_query_rss.Rd b/man/safely_query_rss.Rd
new file mode 100644
index 0000000..3c8bed2
--- /dev/null
+++ b/man/safely_query_rss.Rd
@@ -0,0 +1,19 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/biorecap.R
+\name{safely_query_rss}
+\alias{safely_query_rss}
+\title{Safely query bioRxiv/medRxiv RSS feeds}
+\usage{
+safely_query_rss(subject, server = c("biorxiv", "medrxiv"))
+}
+\arguments{
+\item{subject}{A character vector of valid bioRxiv and/or medRxiv subjects. See \link{subjects}.}
+
+\item{server}{A character vector of either "biorxiv" or "medrxiv".}
+}
+\value{
+A data frame of preprints from bioRxiv and/or medRxiv.
+}
+\description{
+Safely query bioRxiv/medRxiv RSS feeds
+}